Michael Granitzer

:fellow professors

:all professors

Michael Granitzer

Fellow Professor of Information Retrieval & Data Science

:back to overview

Title

Year

Metadata, Structure, or Strategy? A Decomposition of RAG Context Enrichment

2026

Saber Zerhoudi, Michael Granitzer, Jelena Mitrović
Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 57-74
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Jelena Mitrović
Publication date:	09/07/2026
Journal:	Joint European Conference on Machine Learning and Knowledge Discovery in Databases
Pages:	57-74
Publisher:	Springer Nature Switzerland
Description:	Retrieval-augmented generation (RAG) systems increasingly enrich retrieved passages by attaching quality metadata, structuring them into explicit records, and adopting multi-hop retrieval strategies that accumulate evidence across steps. These changes assume that richer context yields better answers, yet existing evaluations cannot test this because they vary all three factors at once. We isolate each factor in a controlled experiment across six benchmarks, four models from three families, and five enrichment levels, totaling over 24,000 evaluated responses. The assumption does not hold. Most enrichment reduces accuracy. Models prompted to use confidence scores comply correctly yet produce worse answers, a gap between utilization and accuracy that no prior work has measured. What determines answer quality is not how much metadata the context carries but whether the model can act on it for the given …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

The Compaction Cliff in Long-Running AI Agent Memory

2026

Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
arXiv preprint arXiv:2608.22752
:read more

Authors:	Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
Publication date:	08/24/2026
Journal:	arXiv preprint arXiv:2608.22752
Description:	A safety rule and an episodic log compete for the same tokens in an AI agent's context. When the budget overflows, both are summarized at the same rate; only the rule needs exact wording to remain enforceable. On 20 production agent configurations, Claude Code's /compact prompt on Sonnet 4.6 preserves 53% of safety rules after one compaction round and 10% after five. We name this the Compaction Cliff. We address it with Knowledge Triage, a framework that classifies each line of an agent's knowledge base by type and routes each type through its own retention policy. Three deterministic operators implement this triage across the three context-management operations: TypeCompact rewrites items in place under per-type fidelity, TypeDecompose partitions a topic too large to compact safely, replicating in-scope safety rules across partitions, and TypeRetrieve fetches items from external storage with in-scope rules pinned ahead of relevance. On five public corpora, TypeCompact preserves 2--4 more safety rules than the strongest single-shot LLM compactor at every ratio, with 96% recall over five rounds. TypeDecompose reaches 0% locality violations against 93% under uniform partitioning. TypeRetrieve reaches 100% recall@50 against 73% for the best single-shot LLM retriever. On three downstream behavioral benchmarks, we outperform the production Sonnet compactor on medical compliance (paired McNemar on preservation, ), the full-policy and hierarchical baselines on retail task pass rate (, ), and the hierarchical compaction on the airline domain (). We release AgentArtifactCorpus …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

When Is Complex Chunking Worth It? A Multi-Objective Evaluation of Chunking Methods at Scale

2026

Laura Caspari, Kanishka Ghosh Dastidar, Michael Dinzinger, Jelena Mitrović, Michael Gra…
arXiv preprint arXiv:2608.16586
:read more

Authors:	Laura Caspari, Kanishka Ghosh Dastidar, Michael Dinzinger, Jelena Mitrović, Michael Granitzer
Publication date:	08/17/2026
Journal:	arXiv preprint arXiv:2608.16586
Description:	Dense retrieval is commonly evaluated on benchmarks that represent each document with a single embedding, even though real-world retrieval systems often index long documents that require chunking. In these settings, the chosen chunking method not only affects retrieval quality, but also indexing throughput, query latency, and memory usage. Prior comparisons of chunking strategies have mainly focused on retrieval performance, leaving operational trade-offs underexplored. To address these issues, we evaluate eight representative chunking strategies across two scalable corpora, three embedding models, and multiple corpus sizes, measuring both retrieval effectiveness and system-level costs. Our results show that computationally expensive methods rarely provide consistent gains over simpler chunking. Instead, the best performing strategy depends on the embedding model, dataset, corpus size, and target retrieval metric. Methods with similar performance can also differ substantially in operational cost, showing that chunking should be seen as a multi-objective design decision.
Resources:	[PDF] from arxiv.org

:view less

Bridging the Gap between Subsampled and Full-Corpus Evaluation

2026

Michael Dinzinger, Kanishka Ghosh Dastidar, Laura Caspari, Jelena Mitrović, Michael Gra…
Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and T…
:read more

Authors:	Michael Dinzinger, Kanishka Ghosh Dastidar, Laura Caspari, Jelena Mitrović, Michael Granitzer
Publication date:	07/25/2026
Journal:	Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)
Pages:	56-61
Description:	Evaluating dense retrieval systems on large corpora is computationally expensive. This poses increasing challenges for scalable experimentation and energy-efficient information retrieval. Recent work on corpus subsampling has shown that evaluation can be conducted on reduced collections consisting of relevant documents and a curated set of high-quality distractors, thereby avoiding the need to encode the full corpus. However, this strategy intentionally disregards the vast majority of non-relevant documents, whose influence on ranking behavior---especially at larger cut-off values---remains insufficiently understood. In this work, we introduce a probabilistic model that accounts for the contribution of the unobserved non-relevant portion of the corpus. We model similarity scores between queries and unseen non-relevant documents as a random variable following an approximate normal distribution and integrate …
Resources:	[PDF] from acm.org

:view less

As We May Search

2026

Saber Zerhoudi, Adam Roegiest, Jelena Mitrović, Michael Granitzer
Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and T…
:read more

Authors:	Saber Zerhoudi, Adam Roegiest, Jelena Mitrović, Michael Granitzer
Publication date:	07/25/2026
Journal:	Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)
Pages:	425-435
Description:	The sensitive information in personal documents, legal files, and medical records is among the most valuable things to search, yet current retrieval-augmented generation systems still require sending content to remote servers. We propose local-first IR, a design philosophy where indexes, models, and inference reside on user devices, treating remote services as optional. This paper makes four contributions: (1) a framework organizing retrieval architectures along three dimensions: privacy and control, capability, and accessibility, (2) experiments on consumer hardware across five benchmarks, scaling from 1K to 1M documents with dense retrieval, BM25, and hybrid fusion. Dense retrieval keeps over 91% nDCG@10 up to 100K documents, with approximate HNSW indexes extending this to 1M with only 2% quality loss; a 7B local language model reaches within 4 points of a cloud baseline on answer quality, (3 …
Cited by:	1
Resources:	[PDF] from acm.org

:view less

Query Performance Prediction under Corpus Growth in Dense Retrieval

2026

Kanishka Ghosh Dastidar, Michael Dinzinger, Laura Caspari, Jelena Mitrović, Michael Gra…
Proceedings of the 49th International ACM SIGIR Conference on Research and Development …
:read more

Authors:	Kanishka Ghosh Dastidar, Michael Dinzinger, Laura Caspari, Jelena Mitrović, Michael Granitzer
Publication date:	07/20/2026
Journal:	Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval
Pages:	3733-3737
Description:	LLM-based chatbots are increasingly augmented with retrieval mechanisms operating over web-scale corpora. Evaluating the effectiveness of these retrieval components is challenging, as explicit relevance judgments are often unavailable. Query performance prediction (QPP) addresses this limitation by providing unsupervised estimates of retrieval effectiveness. However, existing QPP methods assume a static corpus and do not account for the impact of corpus growth on query performance. In this work, we extend the QPP paradigm by studying query performance degradation under corpus inflation in dense retrieval systems. Using tiered corpora with fixed relevance judgments, we analyze how query effectiveness evolves as the corpus (index) size increases and evaluate the ability of established score-based and embedding-based post-retrieval QPP methods to predict such degradation. Our findings show that …
Resources:	[PDF] from acm.org

:view less

NuggetIndex: Governed Atomic Retrieval for Maintainable RAG

2026

Saber Zerhoudi, Michael Granitzer, Jelena Mitrović
Proceedings of the 49th International ACM SIGIR Conference on Research and Development …
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Jelena Mitrović
Publication date:	07/20/2026
Journal:	Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval
Pages:	2286-2296
Description:	Retrieval-augmented generation (RAG) systems are frequently evaluated via fact-based metrics, yet standard implementations retrieve passages or static propositions. This unit mismatch between evaluation and retrieval objects hinders maintenance when corpora evolve and fails to capture superseded facts or source disagreements. We propose NuggetIndex, a retrieval system that stores atomic information units as managed records, so called nuggets. Each record maintains links to evidence, a temporal validity interval, and a lifecycle state. By filtering invalid or deprecated nuggets prior to ranking, the system prevents the inclusion of outdated information. We evaluate the approach using a nuggetized MS MARCO subset, a temporal Wikipedia QA dataset, and a multi-hop QA task. Against passage and unmanaged proposition retrieval baselines, NuggetIndex improves nugget recall by 42%, increases temporal …
Cited by:	2
Resources:	[PDF] from acm.org

:view less

OwlerLite: Scope-and Freshness-Aware Web Retrieval for LLM Assistants

2026

Saber Zerhoudi, Michael Dinzinger, Michael Granitzer, Jelena Mitrovic
Companion Proceedings of the ACM Web Conference 2026, 216-220
:read more

Authors:	Saber Zerhoudi, Michael Dinzinger, Michael Granitzer, Jelena Mitrovic
Publication date:	06/29/2026
Journal:	Companion Proceedings of the ACM Web Conference 2026
Pages:	216-220
Description:	Browser-based language models often use retrieval-augmented generation (RAG) but typically rely on fixed, outdated indices that give users no control over which sources are consulted. This can lead to answers that mix trusted and untrusted content or draw on stale information. We present OwlerLite, a browser-based RAG system that makes user-defined scopes and data freshness central to retrieval. Users define reusable scopes-sets of web pages or sources-and select them when querying. A freshness-aware crawler monitors live pages, uses a semantic change detector to identify meaningful updates, and selectively re-indexes changed content. OwlerLite integrates text relevance, scope choice, and recency into a unified retrieval model. Implemented as a browser extension, it represents a step toward more controllable and trustworthy web assistants.
Cited by:	3
Resources:	[PDF] from acm.org

:view less

Forschungsprojekt DeepWrite

2026

Urs Kramer, Simon Alexander Nonn, Sarah Großkopf, Johann Graf Lambsdorff, Stephan Gesch…
:read more

Authors:	Urs Kramer, Simon Alexander Nonn, Sarah Großkopf, Johann Graf Lambsdorff, Stephan Geschwind, Deborah Voß, Elaheh Alinezhad, Yujin Kang, Michael Granitzer, Lorenz Wendlinger, Abdullah Al Zubaer, Veronika Hackl, Arne Bilzhause
Publication date:	06/22/2026

:view less

Towards multimodal geospatial reasoning: a foundation model approach for disaster detection from social media, news, and weather data

2026

David Hanny, Kanishka Ghosh Dastidar, Marc Wieland, Michael Granitzer, Bernd Resch
Natural Hazards 122 (11), 459
:read more

Authors:	David Hanny, Kanishka Ghosh Dastidar, Marc Wieland, Michael Granitzer, Bernd Resch
Publication date:	06/01/2026
Journal:	Natural Hazards
Volume:	122
Issue:	11
Pages:	459
Publisher:	Springer Netherlands
Description:	The timely detection of disasters is essential for effective emergency response. Traditional satellite-based monitoring provides accurate hazard observations but suffers from acquisition delays and weather-dependent imaging conditions. Therefore, recent research increasingly uses rapidly available digital data such as social media, news, and weather observations. However, most approaches analyse these sources in isolation and lack standardised evaluation. We address this gap using a grid-based framework that quantifies disaster detection accuracy relative to satellite-derived reference data. Within this framework, we introduce a multimodal geospatial reasoning method that employs generative Language Models (LMs) to interpret heterogeneous information. The method integrates Bluesky social media posts, GDELT news headlines, and weather observations through structured prompts and relevance-based …
Cited by:	1
Resources:	[HTML] from springer.com

:view less

The WebKurator. de Platform: Combined Regional and Topical Web Curation

2026

Michael Dinzinger, Natanael Arndt, Ben Böck, Jelena Mitrović, Michael Granitzer
Companion Publication of the 2026 18th ACM Web Science Conference, 40-43
:read more

Authors:	Michael Dinzinger, Natanael Arndt, Ben Böck, Jelena Mitrović, Michael Granitzer
Publication date:	05/26/2026
Journal:	Companion Publication of the 2026 18th ACM Web Science Conference
Pages:	40-43
Description:	The systematic curation of the Web remains a central challenge for national libraries and memory institutions that aim to preserve culturally and regionally relevant content. Existing directory-based approaches such as Curlie implement a predominantly topic-centric, one-dimensional hierarchy, where geographic aspects are intertwined with topical and linguistic categories. To address this limitation, we present WebKurator.de, a collaborative platform for combined regional and topical web curation, initially focused on the German web. WebKurator introduces a two-dimensional curation model that explicitly separates topical categorization and geographic annotation. The system integrates LLM–based topic classification and imprint-based address extraction with geocoding, and supports user suggestions together with moderated review. The platform is bootstrapped from the German Imprints Dataset, a large-scale …
Resources:	[PDF] from acm.org

:view less

GradeLegal: Automated Grading for German Legal Cases

2026

Abdullah Al Zubaer, Lorenz Wendlinger, Simon Alexander Nonn, Michael Granitzer, Jelena …
arXiv preprint arXiv:2605.21076
:read more

Authors:	Abdullah Al Zubaer, Lorenz Wendlinger, Simon Alexander Nonn, Michael Granitzer, Jelena Mitrovic
Publication date:	05/20/2026
Journal:	arXiv preprint arXiv:2605.21076
Description:	Grading German legal exam solutions faces growing volumes and a shortage of qualified graders, delaying feedback and creating a bottleneck. At the same time, it is a high-stakes expert task, since state exam grades strongly influence career outcomes in Germany. Despite this practical relevance, literature lacks systematic studies on effective methods for grading legal exams. To address this gap, we investigate whether large language models (LLMs) can support the automated grading of German legal case solutions in criminal and public law, thereby enabling scalable feedback and student self-testing. We present a systematic evaluation of 27 proprietary and open-source LLMs, benchmarking prompting strategies that incrementally add task-related information, such as a sample solution and a grading rubric. Using quadratic weighted kappa (QWK), reasoning-oriented LLMs can approximate expert grading in public law when given a sample solution and a grading rubric (up to 0.91), compared to 0.60 in criminal law, suggesting a harder grading task in criminal law. Beyond single-model grading, ensembling improves agreement by up to 0.15 over its best member and can offer an alternative to stronger closed-source single models. In addition, our findings suggest that effective prompt design and model selection are necessary for reliable LLM-based grading of legal exams.
Resources:	[PDF] from arxiv.org

:view less

AgentSim: A Platform for Verifiable Agent-Trace Simulation

2026

Saber Zerhoudi, Michael Granitzer, Jelena Mitrovic
arXiv preprint arXiv:2604.26653
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Jelena Mitrovic
Publication date:	04/29/2026
Journal:	arXiv preprint arXiv:2604.26653
Description:	Training trustworthy agentic LLMs requires data that shows the grounded reasoning process, not just the final answer. Existing datasets fall short: question-answering data is outcome-only, chain-of-thought data is not tied to specific documents, and web-agent datasets track interface actions rather than the core retrieval and synthesis steps of a RAG workflow. We introduce AgentSim, an open-source platform for simulating RAG agents. It generates verifiable, stepwise traces of agent reasoning over any document collection. AgentSim uses a policy to ensure the agent widely explores the document set. It combines a multi-model validation pipeline with an active human-in-the-loop process. This approach focuses human effort on difficult steps where models disagree. Using AgentSim, we construct and release the Agent-Trace Corpus (ATC), a large collection of grounded reasoning trajectories spanning three established IR benchmarks. We make three contributions: (1) the AgentSim platform with two mechanisms, Corpus-Aware Seeding and Active Validation, that improve trace diversity and quality; (2) the Agent-Trace Corpus (ATC), over 103,000 verifiable reasoning steps spanning three IR benchmarks, with 100% grounding rate on substantive answers; and (3) a comparative behavioral analysis revealing systematic differences in how state-of-the-art models approach information seeking. Platform, toolkit, and corpus are publicly available.
Cited by:	2
Resources:	[PDF] from arxiv.org

:view less

IIRSim Studio: A Dashboard for User Simulation

2026

Saber Zerhoudi, Adam Roegiest, Michael Granitzer
arXiv preprint arXiv:2604.23406
:read more

Authors:	Saber Zerhoudi, Adam Roegiest, Michael Granitzer
Publication date:	04/25/2026
Journal:	arXiv preprint arXiv:2604.23406
Description:	User simulation is a valuable methodology for evaluation in Information Retrieval (IR), enabling low-cost experimentation and counterfactual analysis. However, existing simulation frameworks are primarily code-centric libraries that require substantial setup effort, which limits adoption and hinders reproducibility. The bottleneck is not the simulation engines themselves, but the lack of infrastructure connecting experiment design, execution, and sharing into a single verifiable workflow. This paper introduces IIRSim Studio, a web-based workbench that addresses this gap through four contributions: (1) a visual environment for composing simulation pipelines on top of simulation frameworks, serving both novices learning simulation concepts and experts piloting large-scale experiments; (2) a component lifecycle that supports authoring, versioning, and sharing custom simulation components through Git-backed storage and runtime injection; (3) a provenance model based on experiment bundles and environment templates that makes the scope of replication explicit; and (4) a shared-task workflow, demonstrated through the re-deployment of a Sim4IA micro-task. IIRSim Studio is available as a hosted service and as a portable containerized deployment.
Resources:	[PDF] from arxiv.org

:view less

Creating Specialized RAG-Based Search Engines Using the Open Web Index

2026

Alexander Nussbaumer, Michael Dinzinger, Sebastian Heineking, Gijs Hendriksen, Felix Ho…
European Conference on Information Retrieval, 206-211
:read more

Authors:	Alexander Nussbaumer, Michael Dinzinger, Sebastian Heineking, Gijs Hendriksen, Felix Holz, Saber Zerhoudi, Martin Potthast, Michael Granitzer
Publication date:	03/24/2026
Journal:	European Conference on Information Retrieval
Pages:	206-211
Publisher:	Springer Nature Switzerland
Description:	This paper presents a concept and supporting technology for building RAG-based specialized search engines using open-source frameworks and open web data. The Open Web Index (OWI) provides openly accessible web data, while the modular MOSAIC framework is designed to integrate topical OWI partitions obtained to create search applications tailored to specific use cases. MOSAIC-RAG extends this framework with features based on Large Language Models (LLM), such as summarization or re-ranking. Using this infrastructure, special-purpose and domain-specific search applications can easily be developed and experimented with. For demonstration purposes, we present three example applications in the topical domains of science, health, and arts.

:view less

Beyond the Click: A Framework for Inferring Cognitive Traces in Search

2026

Saber Zerhoudi, Michael Granitzer
European Conference on Information Retrieval, 626-640
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	03/24/2026
Journal:	European Conference on Information Retrieval
Pages:	626-640
Publisher:	Springer Nature Switzerland
Description:	User simulators are essential for evaluating search systems, but they primarily copy user actions without understanding the underlying thought process. This gap exists since large-scale interaction logs record what users do, but not what they might be thinking or feeling, such as confusion or satisfaction. To solve this problem, we present a framework to infer cognitive traces from behavior logs. Our method uses a multi-agent system grounded in Information Foraging Theory (IFT) and human expert judgment. These traces improve model performance on tasks like forecasting session outcomes and user struggle recovery. We release a collection of annotations for several public datasets, including AOL and Stack Overflow, and an open-source tool that allows researchers to apply our method to their own data (https://github.com/searchsim-org/cognitive-traces). This work provides the tools and data needed to build more …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

The Third International Workshop on Open Web Search (WOWS)

2026

Laura Caspari, Maik Fröbe, Sebastian Heineking, Michael Granitzer, Gijs Hendriksen, Djo…
European Conference on Information Retrieval, 231-238
:read more

Authors:	Laura Caspari, Maik Fröbe, Sebastian Heineking, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra, Martin Potthast, Arjen P de Vries, Saber Zerhoudi
Publication date:	03/24/2026
Journal:	European Conference on Information Retrieval
Pages:	231-238
Publisher:	Springer Nature Switzerland
Description:	We organize the third International Workshop on Open Web Search (WOWS) at ECIR 2026 with two calls for contributions: The first call targets scientific contributions on collaborative search engine building, including crawling, deployment, evaluation, and use of the web as a resource by researchers and innovators. The second call is for the WOWS-Eval shared task, which aims to gain practical experience with collaborative, cooperative search engine evaluation by focusing on using the Open Web Index (OWI) for (RAG) retrieval experiments.
Resources:	[PDF] from djoerdhiemstra.com

:view less

CoRECT: A framework for evaluating embedding compression techniques at scale

2026

Laura Caspari, Michael Dinzinger, Kanishka Ghosh Dastidar, Christofer Fellicious, Jelen…
European Conference on Information Retrieval, 383-398
:read more

Authors:	Laura Caspari, Michael Dinzinger, Kanishka Ghosh Dastidar, Christofer Fellicious, Jelena Mitrović, Michael Granitzer
Publication date:	03/24/2026
Journal:	European Conference on Information Retrieval
Pages:	383-398
Publisher:	Springer Nature Switzerland
Description:	Dense retrieval systems have proven to be effective across various benchmarks, but require substantial memory to store large search indices. Recent advances in embedding compression show that index sizes can be greatly reduced with minimal loss in ranking quality. However, existing studies often overlook the role of corpus complexity – a critical factor, as recent work shows that both corpus size and document length strongly affect dense retrieval performance. In this paper, we introduce CoRECT (Controlled Retrieval Evaluation of Compression Techniques), a framework for large-scale evaluation of embedding compression methods, supported by a newly curated dataset collection. To demonstrate its utility, we benchmark eight representative types of compression methods. Notably, we show that non-learned compression achieves substantial index size reduction, even on up to 100M passages, with …
Cited by:	6
Resources:	[PDF] from arxiv.org

:view less

From SERPs to Agents: A Platform for Comparative Studies of Information Interaction

2026

Saber Zerhoudi, Michael Granitzer
Proceedings of the 2026 Conference on Human Information Interaction and Retrieval, 122-126
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	03/22/2026
Journal:	Proceedings of the 2026 Conference on Human Information Interaction and Retrieval
Pages:	122-126
Description:	The diversification of information access systems, from RAG to autonomous agents, creates a critical need for comparative user studies. However, the technical overhead to deploy and manage these distinct systems is a major barrier. We present UXLab ,, an open-source system for web-based user studies that addresses this challenge. Its core is a web-based dashboard enabling the complete, no-code configuration of complex experimental designs. Researchers can visually manage the full study, from recruitment to comparing backends like traditional search, vector databases, and LLMs. We demonstrate UXLab’s value via a micro case study comparing user behavior with RAG versus an autonomous agent. UXLab allows researchers to focus on experimental design and analysis, supporting future multi-modal interaction research.
Resources:	[HTML] from acm.org

:view less

In-browser agents for search assistance

2026

Saber Zerhoudi, Michael Granitzer
Proceedings of the 2026 Conference on Human Information Interaction and Retrieval, 345-349
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	03/22/2026
Journal:	Proceedings of the 2026 Conference on Human Information Interaction and Retrieval
Pages:	345-349
Description:	A fundamental tension exists between the demand for sophisticated AI assistance in web search and the need for user data privacy. Current centralized models require users to transmit sensitive browsing data to external services, which limits user control. In this paper, we present a browser extension that provides a viable in-browser alternative. We introduce a hybrid architecture that functions entirely on the client side, combining two components: (1) an adaptive probabilistic model that learns a user’s behavioral policy from direct feedback, and (2) a Small Language Model (SLM), running in the browser, which is grounded by the probabilistic model to generate context-aware suggestions. To evaluate this approach, we conducted a three-week longitudinal user study with 18 participants. Our results show that this privacy-preserving approach is highly effective at adapting to individual user behavior, leading to …
Cited by:	1
Resources:	[PDF] from acm.org

:view less

Toward Generalized Cross-Lingual Hateful Language Detection with Web-Scale Data and Ensemble LLM Annotations

2026

Dang H Dang, Jelena Mitrovi, Michael Granitzer
arXiv preprint arXiv:2604.09625
:read more

Authors:	Dang H Dang, Jelena Mitrovi, Michael Granitzer
Publication date:	03/18/2026
Journal:	arXiv preprint arXiv:2604.09625
Description:	We study whether large-scale unlabelled web data and LLM-based synthetic annotations can improve multilingual hate speech detection. Starting from texts crawled via OpenWebSearch.eu~(OWS) in four languages (English, German, Spanish, Vietnamese), we pursue two complementary strategies. First, we apply continued pre-training to BERT models by continuing masked language modelling on unlabelled OWS texts before supervised fine-tuning, and show that this yields an average macro-F1 gain of approximately 3% over standard baselines across sixteen benchmarks, with stronger gains in low-resource settings. Second, we use four open-source LLMs (Mistral-7B, Llama3.1-8B, Gemma2-9B, Qwen2.5-14B) to produce synthetic annotations through three ensemble strategies: mean averaging, majority voting, and a LightGBM meta-learner. The LightGBM ensemble consistently outperforms the other strategies. Fine-tuning on these synthetic labels substantially benefits a small model (Llama3.2-1B: +11% pooled F1), but provides only a modest gain for the larger Qwen2.5-14B (+0.6%). Our results indicate that the combination of web-scale unlabelled data and LLM-ensemble annotations is the most valuable for smaller models and low-resource languages.
Resources:	[PDF] from arxiv.org

:view less

Form Without Function: Agent Social Behavior in the Moltbook Network

2026

Saber Zerhoudi, Kanishka Ghosh Dastidar, Felix Klement, Artur Romazanov, Andreas Einwil…
arXiv preprint arXiv:2604.13052
:read more

Authors:	Saber Zerhoudi, Kanishka Ghosh Dastidar, Felix Klement, Artur Romazanov, Andreas Einwiller, Dang H Dang, Michael Dinzinger, Michael Granitzer, Annette Hautli-Janisz, Stefan Katzenbeisser, Florian Lemmerich, Jelena Mitrovic
Publication date:	03/17/2026
Journal:	arXiv preprint arXiv:2604.13052
Description:	Moltbook is a social network where every participant is an AI agent. We analyze 1,312,238 posts, 6.7~million comments, and over 120,000 agent profiles across 5,400 communities, collected over 40 days (January 27 to March 9, 2026). We evaluate the platform through three layers. At the interaction layer, 91.4% of post authors never return to their own threads, 85.6% of conversations are flat (no reply ever receives a reply), the median time-to-first-comment is 55 seconds, and 97.3% of comments receive zero upvotes. Interaction reciprocity is 3.3%, compared to 22-60% on human platforms. An argumentation analysis finds that 64.6% of comment-to-post relations carry no argumentative connection. At the content layer, 97.9% of agents never post in a community matching their bio, 92.5% of communities contain every topic in roughly equal proportions, and over 80% of shared URLs point to the platform's own infrastructure. At the instruction layer, we use 41 Wayback Machine snapshots to identify six instruction changes during the observation window. Hard constraints (rate limit, content filters) produce immediate behavioral shifts. Soft guidance (``upvote good posts'', ``stay on topic'') is ignored until it becomes an explicit step in the executable checklist. The platform also poses technological risks. We document credential leaks (API keys, JWT tokens), 12,470 unique Ethereum addresses with 3,529 confirmed transaction histories, and attack discourse ranging from template-based SSH brute-forcing to multi-agent offensive security architectures. These persist unmoderated because the quality-filtering mechanisms are themselves non-functional …
Cited by:	2
Resources:	[PDF] from arxiv.org

:view less

Behind the Prompt: The Agent-User Problem in Information Retrieval

2026

Saber Zerhoudi, Michael Granitzer, Dang Hai Dang, Jelena Mitrovic, Florian Lemmerich, A…
arXiv preprint arXiv:2603.03630
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Dang Hai Dang, Jelena Mitrovic, Florian Lemmerich, Annette Hautli-Janisz, Stefan Katzenbeisser, Kanishka Ghosh Dastidar
Publication date:	03/04/2026
Journal:	arXiv preprint arXiv:2603.03630
Description:	User models in information retrieval rest on a foundational assumption that observed behavior reveals intent. This assumption collapses when the user is an AI agent privately configured by a human operator. For any action an agent takes, a hidden instruction could have produced identical output - making intent non-identifiable at the individual level. This is not a detection problem awaiting better tools; it is a structural property of any system where humans configure agents behind closed doors. We investigate the agent-user problem through a large-scale corpus from an agent-native social platform: 370K posts from 47K agents across 4K communities. Our findings are threefold: (1) individual agent actions cannot be classified as autonomous or operator-directed from observables; (2) population-level platform signals still separate agents into meaningful quality tiers, but a click model trained on agent interactions degrades steadily (-8.5% AUC) as lower-quality agents enter training data; (3) cross-community capability references spread endemically ( 1.26-3.53) and resist suppression even under aggressive modeled intervention. For retrieval systems, the question is no longer whether agent users will arrive, but whether models built on human-intent assumptions will survive their presence.
Cited by:	3
Resources:	[PDF] from arxiv.org

:view less

WebFAQ 2.0: A Multilingual QA Dataset with Mined Hard Negatives for Dense Retrieval

2026

Michael Dinzinger, Laura Caspari, Ali Salman, Irvin Topi, Jelena Mitrović, Michael Gran…
arXiv preprint arXiv:2602.17327
:read more

Authors:	Michael Dinzinger, Laura Caspari, Ali Salman, Irvin Topi, Jelena Mitrović, Michael Granitzer
Publication date:	02/19/2026
Journal:	arXiv preprint arXiv:2602.17327
Description:	We introduce WebFAQ 2.0, a new version of the WebFAQ dataset, containing 198 million FAQ-based natural question-answer pairs across 108 languages. Compared to the previous version, it significantly expands multilingual coverage and the number of bilingual aligned QA pairs to over 14.3M, making it the largest FAQ-based resource. Unlike the original release, WebFAQ 2.0 uses a novel data collection strategy that directly crawls and extracts relevant web content, resulting in a substantially more diverse and multilingual dataset with richer context through page titles and descriptions. In response to community feedback, we also release a hard negatives dataset for training dense retrievers, with 1.25M queries across 20 languages. These hard negatives were mined using a two-stage retrieval pipeline and include cross-encoder scores for 200 negatives per query. We further show how this resource enables two primary fine-tuning strategies for dense retrievers: Contrastive Learning with MultipleNegativesRanking loss, and Knowledge Distillation with MarginMSE loss. WebFAQ 2.0 is not a static resource but part of a long-term effort. Since late 2025, structured FAQs are being regularly released through the Open Web Index, enabling continuous expansion and refinement. We publish the datasets and training scripts to facilitate further research in multilingual and cross-lingual IR. The dataset itself and all related resources are publicly available on GitHub and HuggingFace.
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

Robust Generalizable Heterogeneous Legal Link Prediction

2026

Lorenz Wendlinger, Simon Alexander Nonn, Abdullah Al Zubaer, Michael Granitzer
arXiv preprint arXiv:2602.04812
:read more

Authors:	Lorenz Wendlinger, Simon Alexander Nonn, Abdullah Al Zubaer, Michael Granitzer
Publication date:	02/04/2026
Journal:	arXiv preprint arXiv:2602.04812
Description:	Recent work has applied link prediction to large heterogeneous legal citation networks new{with rich meta-features}. We find that this approach can be improved by including edge dropout and feature concatenation for the learning of more robust representations, which reduces error rates by up to 45%. We also propose an approach based on multilingual node features with an improved asymmetric decoder for compatibility, which allows us to generalize and extend the prediction to more, geographically and linguistically disjoint, data from New Zealand. Our adaptations also improve inductive transferability between these disjoint legal systems.
Resources:	[PDF] from arxiv.org

:view less

A multimodal GeoAI approach to combining text with spatiotemporal features for enhanced relevance classification of social media posts in disaster response

2026

David Hanny, Sebastian Schmidt, Shaily Gandhi, Michael Granitzer, Bernd Resch
Big Earth Data 10 (1), 258-302
:read more

Authors:	David Hanny, Sebastian Schmidt, Shaily Gandhi, Michael Granitzer, Bernd Resch
Publication date:	01/02/2026
Journal:	Big Earth Data
Volume:	10
Issue:	1
Pages:	258-302
Publisher:	Taylor & Francis
Description:	Geo-referenced social media data supports disaster management by offering real-time insights through user-generated content. To identify critical information amid high volumes of noise, classifying the relevance of posts is essential. Most existing methods primarily use textual features, neglecting spatial and temporal context despite its importance in determining relevance. This study proposes a multimodal approach that integrates text with spatiotemporal features for relevance classification of geo-referenced social media posts. We evaluate our method on 4,574 manually labelled posts from five disasters: the 2020 California wildfires, 2021 Ahr Valley floods, 2023 Chile wildfires, 2023 Turkey earthquake and 2023 Emilia-Romagna floods. Labels were assigned based on text, geographic location and time. Our spatiotemporal features include proximity to disaster impact sites, local co-occurrences with disaster …
Cited by:	11
Resources:	[PDF] from tandfonline.com

:view less

Benevolent Dictators? On LLM Agent Behavior in Dictator Games

2025

Andreas Einwiller, Kanishka Ghosh Dastidar, Artur Romazanov, Annette Hautli-Janisz, Mic…
arXiv preprint arXiv:2511.08721
:read more

Authors:	Andreas Einwiller, Kanishka Ghosh Dastidar, Artur Romazanov, Annette Hautli-Janisz, Michael Granitzer, Florian Lemmerich
Publication date:	11/11/2025
Journal:	arXiv preprint arXiv:2511.08721
Description:	In behavioral sciences, experiments such as the ultimatum game are conducted to assess preferences for fairness or self-interest of study participants. In the dictator game, a simplified version of the ultimatum game where only one of two players makes a single decision, the dictator unilaterally decides how to split a fixed sum of money between themselves and the other player. Although recent studies have explored behavioral patterns of AI agents based on Large Language Models (LLMs) instructed to adopt different personas, we question the robustness of these results. In particular, many of these studies overlook the role of the system prompt - the underlying instructions that shape the model's behavior - and do not account for how sensitive results can be to slight changes in prompts. However, a robust baseline is essential when studying highly complex behavioral aspects of LLMs. To overcome previous limitations, we propose the LLM agent behavior study (LLM-ABS) framework to (i) explore how different system prompts influence model behavior, (ii) get more reliable insights into agent preferences by using neutral prompt variations, and (iii) analyze linguistic features in responses to open-ended instructions by LLM agents to better understand the reasoning behind their behavior. We found that agents often exhibit a strong preference for fairness, as well as a significant impact of the system prompt on their behavior. From a linguistic perspective, we identify that models express their responses differently. Although prompt sensitivity remains a persistent challenge, our proposed framework demonstrates a robust foundation for LLM agent …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

Uxsim: Towards a hybrid user search simulation

2025

Saber Zerhoudi, Michael Granitzer
Proceedings of the 34th ACM International Conference on Information and Knowledge Manag…
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	11/10/2025
Journal:	Proceedings of the 34th ACM International Conference on Information and Knowledge Management
Pages:	6570-6574
Description:	Simulating nuanced user experiences within complex interactive search systems poses distinct challenge for traditional methodologies, which often rely on static user proxies or, more recently, on standalone large language model (LLM) agents that may lack deep, verifiable grounding. The true dynamism and personalization inherent in human-computer interaction demand a more integrated approach. This work introduces UXSim. https://searchsim.org/uxsim, a novel framework that integrates both approaches. It leverages grounded data from traditional simulators to inform and constrain the reasoning of an adaptive LLM agent. This synthesis enables more accurate and dynamic simulations of user behavior while also providing a pathway for the explainable validation of the underlying cognitive processes.
Cited by:	6
Resources:	[PDF] from acm.org

:view less

Compressed concatenation of small embedding models

2025

Ben Ayad Mohamed Ayoub, Michael Dinzinger, Kanishka Ghosh Dastidar, Jelena Mitrović, Mi…
Proceedings of the 34th ACM International Conference on Information and Knowledge Manag…
:read more

Authors:	Ben Ayad Mohamed Ayoub, Michael Dinzinger, Kanishka Ghosh Dastidar, Jelena Mitrović, Michael Granitzer
Publication date:	11/10/2025
Journal:	Proceedings of the 34th ACM International Conference on Information and Knowledge Management
Pages:	4613-4617
Description:	Embedding models are central to dense retrieval, semantic search, and recommendation systems, but their size often makes them impractical to deploy in resource-constrained environments such as browsers or edge devices. While smaller embedding models offer practical advantages, they typically underperform compared to their larger counterparts. To bridge this gap, we demonstrate that concatenating the raw embedding vectors of multiple small models can outperform a single larger baseline on standard retrieval benchmarks. To overcome the resulting high dimensionality of naive concatenation, we introduce a lightweight unified decoder trained with a Matryoshka Representation Learning (MRL) loss. This decoder maps the high-dimensional joint representation to a low-dimensional space, preserving most of the original performance without fine-tuning the base models. We also show that while …
Cited by:	2
Resources:	[PDF] from acm.org

:view less

Comparative Study of Language Models and Prompt Paradigms in Short Answer Grading

2025

Abdullah Al Zubaer, Stephan Geschwind, Deborah Voss, Lorenz Wendlinger, Johann Graf Lam…
2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI), …
:read more

Authors:	Abdullah Al Zubaer, Stephan Geschwind, Deborah Voss, Lorenz Wendlinger, Johann Graf Lambsdorff, Michael Granitzer, Jelena Mitrović
Publication date:	11/03/2025
Journal:	2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI)
Pages:	321-329
Publisher:	IEEE
Description:	Grades deliver important quantitative feedback on student learning, but grading can be time-consuming for large student cohorts. Recent advances in large language models (LLMs) offer promising opportunities for automating this process. Still, a systematic examination of different models and prompting paradigms remains lacking. To fill this gap, we compare the performance of 23 closed-source and open-source LLMs, testing the best-performing six across five different prompt paradigms including zero-shot, pre-written sample solutions with an integrated grading rubric, few-shot examples, and retrievalaugmented generation (RAG) based on textbook excerpts. We hence investigate the extent to which external information is necessary for reliable grading outcomes. As a benchmark, we include four fine-tuned encoder models. Our study focuses on German-language undergraduate economics courses, using an …
Cited by:	1

:view less

Cross-Domain and Multilingual Analysis of Retrieval Augmented Generation on Ontologies

2025

Ramona Kühn, Jelena Mitrović, Lorenz Wendlinger, Michael Granitzer
2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI), …
:read more

Authors:	Ramona Kühn, Jelena Mitrović, Lorenz Wendlinger, Michael Granitzer
Publication date:	11/03/2025
Journal:	2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI)
Pages:	1396-1403
Publisher:	IEEE
Description:	Retrieval-Augmented Generation (RAG) extends LLMs with external knowledge, typically from continuous text sources. In this work, we explore the application of RAG to ontologies to support ontology engineers and users in querying structured knowledge. To demonstrate the generalizability of our approach, we evaluate RAG on three ontologies from the domains of video games, IoT infrastructure, and rhetorical figures. Two of the ontologies are in English, while one is in German. Our study compares chunking strategies, chunk sizes, and the use of rerankers. We observe that advanced chunking techniques offer limited performance gains due to the inherently modular structure of ontologies. Additionally, we compare RAG on ontological document structure with the same information in a continuous format and reveal that the performance of RAG is not negatively affected by the ontological document structure. An …

:view less

Joint Learning for Efficient German Argument Mining

2025

Lorenz Wendlinger, Ramona Kühn, Jelena Mitrović, Michael Granitzer
2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI), …
:read more

Authors:	Lorenz Wendlinger, Ramona Kühn, Jelena Mitrović, Michael Granitzer
Publication date:	11/03/2025
Journal:	2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI)
Pages:	770-777
Publisher:	IEEE
Description:	We propose a lightweight and adaptable model tailored to German argument mining in two dissimilar domains. It offers improvements over the state-of-the-art, as well as over recent LLM approaches. Specifically, we achieve an 11.5 % absolute improvement across all classes of the SPWSLE dataset as well as 3.5% absolute detection performance boost across all datasets of the CIMT collection. We show that for both tasks, the hitherto disjoint steps of argument detection followed by argument classification can be merged with no loss in quality, resulting in a greater than training and inference speed-up. For the naive baseline, this joint scheme also significantly improves results. For CIMT we propose realistic evaluation that incorporates actual detection results instead of simulating perfect detection. We also investigate the transferability between and to new datasets and find it to be higher than in previous work …

:view less

GPT-4 shows comparable performance to human examiners in ranking open-text answers

2025

Abdullah Al Zubaer, Michael Granitzer, Stephan Geschwind, Johann Graf Lambsdorff, Debor…
Scientific Reports 15 (1), 35045
:read more

Authors:	Abdullah Al Zubaer, Michael Granitzer, Stephan Geschwind, Johann Graf Lambsdorff, Deborah Voss
Publication date:	10/08/2025
Journal:	Scientific Reports
Volume:	15
Issue:	1
Pages:	35045
Publisher:	Nature Publishing Group UK
Description:	Can GPT-4 replace human examiners? To address this question we explore the performance of GPT-4 as an examiner of answers to open-text questions. We formulate questions and sample solutions in the field of macroeconomics and collect answers from cohorts of undergraduate students. We then conduct a fair competition between GPT-4 and human experts, employing their expertise to assess the quality of the answers. We observe that the substitution of GPT-4 for a human examiner does not decrease inter-rater reliability on tasks that rank the quality of answers. We run checks on potential biases (whether GPT-4 prefers AI-generated or lengthy answers). We find no consistent evidence of such biases. Our findings are robust to tilting the competition to one side’s advantage, by using inferior or advanced prompting strategies. Our results are more attenuated on tasks where GPT-4 assigns points to student …
Cited by:	7
Resources:	[HTML] from nature.com

:view less

ARCHITECTING THE DATASTORE FOR THE URL FRONTIER OF OPENWEBSEARCH. EU

2025

Noor A Fathima, M Dinzinger, M Granitzer, A Wagner
7th International Open Search Symposium# ossym2025, 1
:read more

Authors:	Noor A Fathima, M Dinzinger, M Granitzer, A Wagner
Publication date:	10/08/2025
Journal:	7th International Open Search Symposium# ossym2025
Pages:	1
Description:	This paper presents the architectural evolution of the URL Frontier datastore within the OpenWebSearch. eu initiative [2], transitioning from an OpenSearch-based prototype [3] to a high-throughput ScyllaDB deployment [4]. Motivated by the need for low-latency, write-optimized infrastructure to support continuous web crawling, we conducted a structured evaluation comparing OpenSearch, HBase [6], Cassandra [7], and ScyllaDB across performance, scalability, operational complexity, and infrastructure compatibility. Our findings identified ScyllaDB as the most suitable datastore due to its shard-per-core design using the Seastar framework [8], SSD optimization, and minimal maintenance overhead. We detail the deployment process using rootless containers managed via Podman [9] and secured through Puppet-managed nftables [10, 11], as well as the integration with Scylla Manager for future cluster scaling [5].
Resources:	[PDF] from cern.ch

:view less

Compressed Concatenation of Small Embedding Models

2025

Mohamed Ayoub Ben Ayad, Michael Dinzinger, Kanishka Ghosh Dastidar, Jelena Mitrovic, Mi…
arXiv e-prints, arXiv: 2510.04626
:read more

Authors:	Mohamed Ayoub Ben Ayad, Michael Dinzinger, Kanishka Ghosh Dastidar, Jelena Mitrovic, Michael Granitzer
Publication date:	10/01/2025
Journal:	arXiv e-prints
Pages:	arXiv: 2510.04626
Description:	Embedding models are central to dense retrieval, semantic search, and recommendation systems, but their size often makes them impractical to deploy in resource-constrained environments such as browsers or edge devices. While smaller embedding models offer practical advantages, they typically underperform compared to their larger counterparts. To bridge this gap, we demonstrate that concatenating the raw embedding vectors of multiple small models can outperform a single larger baseline on standard retrieval benchmarks. To overcome the resulting high dimensionality of naive concatenation, we introduce a lightweight unified decoder trained with a Matryoshka Representation Learning (MRL) loss. This decoder maps the high-dimensional joint representation to a low-dimensional space, preserving most of the original performance without fine-tuning the base models. We also show that while …

:view less

On the Suitability of pre-trained foundational LLMs for Analysis in German Legal Education

2025

Lorenz Wendlinger, Christian Braun, Abdullah Al Zubaer, Simon Alexander Nonn, Sarah Gro…
International Artificial Intelligence Symposium, 159-174
:read more

Authors:	Lorenz Wendlinger, Christian Braun, Abdullah Al Zubaer, Simon Alexander Nonn, Sarah Großkopf, Christofer Fellicious, Michael Granitzer
Publication date:	09/21/2025
Journal:	International Artificial Intelligence Symposium
Pages:	159-174
Publisher:	Springer Nature Switzerland
Description:	We show that current open-source foundational LLMs possess instruction capability and German legal background knowledge that is sufficient for some legal analysis in an educational context. However, model capability degrades in very specific tasks, such as the classification of “Gutachtenstil” appraisal style components, or with complex contexts, such as complete legal opinions. Even with extended context and effective prompting strategies, they cannot match the Bag-of-Words baseline. To combat this, we introduce a Retrieval Augmented Generation based prompt example selection method that substantially improves predictions in high data availability scenarios. We further evaluate the performance of pre-trained LLMs on two standard tasks for argument mining and automated essay scoring and find it to be more adequate. Throughout, pre-trained LLMs improve upon the baseline in scenarios with little or no …
Cited by:	3
Resources:	[PDF] from arxiv.org

:view less

Computational approaches to the detection of lesser-known rhetorical figures: A systematic survey and research challenges

2025

Ramona Kühn, Jelena Mitrović, Michael Granitzer
ACM Computing Surveys 58 (2), 1-36
:read more

Authors:	Ramona Kühn, Jelena Mitrović, Michael Granitzer
Publication date:	09/09/2025
Journal:	ACM Computing Surveys
Volume:	58
Issue:	2
Pages:	1-36
Publisher:	ACM
Description:	Rhetorical figures play a major role in everyday communication, making text and speech more interesting, memorable, or persuasive through their association between form and meaning. Computational detection of rhetorical figures plays an important part in thorough understanding of complex communication patterns. In this survey, we provide a comprehensive overview of computational approaches to lesser-known rhetorical figures. We explore the linguistic and computational perspectives on rhetorical figures and highlight their significance in the field of Natural Language Processing. We present different figures in detail and investigate datasets, definitions, rhetorical functions, and detection approaches. We identify challenges such as dataset scarcity, language limitations, and reliance on rule-based methods.
Cited by:	7
Resources:	[PDF] from acm.org Full View

:view less

The Missing Link: Joint Legal Citation Prediction Using Heterogeneous Graph Enrichment

2025

Lorenz Wendlinger, Simon Alexander Nonn, Abdullah Al Zubaer, Michael Granitzer
International Conference on Database and Expert Systems Applications, 197-211
:read more

Authors:	Lorenz Wendlinger, Simon Alexander Nonn, Abdullah Al Zubaer, Michael Granitzer
Publication date:	08/18/2025
Journal:	International Conference on Database and Expert Systems Applications
Pages:	197-211
Publisher:	Springer Nature Switzerland
Description:	Legal systems heavily rely on cross-citations of legal norms as well as previous court decisions. Practitioners, novices and legal AI systems need access to these relevant data to inform appraisals and judgments. We propose a Graph-Neural-Network (GNN) link prediction model that can identify Case-Law and Case-Case citations with high proficiency through fusion of semantic and topological information. We introduce adapted relational graph convolutions operating on an extended and enriched version of the original citation graph that allow the topological integration of semantic meta-information. This further improves prediction by 3.1 points of average precision and by 8.5 points in data sparsity as well as showing robust performance over time and in challenging fully inductive prediction. Jointly learning and predicting case and norm citations achieves a large synergistic effect that improves case citation …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

Webfaq: A multilingual collection of natural q&a datasets for dense retrieval

2025

Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrović, Michael Gra…
Proceedings of the 48th International ACM SIGIR Conference on Research and Development …
:read more

Authors:	Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrović, Michael Granitzer
Publication date:	07/13/2025
Journal:	Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval
Pages:	3802-3811
Description:	We present WebFAQ, a large-scale collection of open-domain question answering datasets derived from FAQ-style schema.org annotations. In total, the data collection consists of 96 million natural question-answer (QA) pairs across 75 languages, including 47 million (49%) non-English samples. WebFAQ further serves as the foundation for 49 monolingual retrieval benchmarks with a total size of 11.2 million QA pairs (5.9 million non-English). These datasets are carefully curated through refined filtering and near-duplicate detection, yielding high-quality resources for training and evaluating multilingual dense retrieval models. To empirically confirm WebFAQ's efficacy, we use the collected QAs to fine-tune an in-domain pretrained XLM-RoBERTa model. Through this process of dataset-specific fine-tuning, the model achieves significant retrieval performance gains, which generalize - beyond WebFAQ - to other …
Cited by:	17
Resources:	[PDF] from acm.org

:view less

Class-Based Feature Knowledge Distillation

2025

Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Enhancing LLM Performance: Efficacy, Fine-Tuning, and Inference Techniques, 119-130
:read more

Authors:	Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Publication date:	07/05/2025
Journal:	Enhancing LLM Performance: Efficacy, Fine-Tuning, and Inference Techniques
Pages:	119-130
Publisher:	Springer Nature Switzerland
Description:	Knowledge Distillation (KD) is an effective technique for compressing large language models through the teacher-student framework without compromising accuracy. Once a large model is compressed to a lightweight one, it can be stored within limited disk space and trained at higher speed with fewer computational resources. Previous work on feature distillation mainly applied an exact matching between the hidden representations of the student and the teacher. However, since the student has a lower capacity compared to the teacher, it may struggle to mimic its exact hidden representations. This leads to a large discrepancy between their features as demonstrated in preceding research. Therefore, we propose intra-class similarity-guided feature distillation, a novel approach to make the task easier for the student. In this chapter, we map each sample representation by the student to its k-nearest neighbor …
Cited by:	1

:view less

Report on the 2nd search futures workshop at ECIR 2025

2025

Charles LA Clarke, Paul Kantor, Adam Roegiest, Johanne R Trippas, Zhaochun Ren, Maria S…
ACM SIGIR Forum 59 (1), 1-28
:read more

Authors:	Charles LA Clarke, Paul Kantor, Adam Roegiest, Johanne R Trippas, Zhaochun Ren, Maria Sofia Bucarelli, Xiao Fu, Yixing Fan, Michael Granitzer, David Graus, Maria Heuss, Jaap Kamps, Yibin Lei, Andrew Parry, Damiaan Reijnaers, Maarten De Rijke, Siddharth AK Singh, Yubao Tang, Suzan Verberne, Jonas Wallat, Yumeng Wang, Chen Xu, Andrew Yates, Saber Zerhoudi, Jujia Zhao
Publication date:	06/01/2025
Journal:	ACM SIGIR Forum
Volume:	59
Issue:	1
Pages:	1-28
Publisher:	ACM
Description:	The Second Search Futures Workshop, in conjunction with the Forty-seventh European Conference on Information Retrieval (ECIR) 2025, looked into the future of search to ask questions such as: • How can we navigate data privacy in large language model (LLM)-based information retrieval (IR)? • How can we implement agentic IR for proactive knowledge synthesis? • How do we ensure trustworthy information access beyond citations in the age of language models? • How does deep search transition from matching to reasoning? • What is meant by information semantics, knowledge representation, and natural language in a world of LLM-powered search? • What are serendipity engines, and how do they explore proactive web search via LLM agents, retrieval augmented generation (RAG), and simulated user feedback? The second edition of the workshop opened with ten lightning talks from a …
Cited by:	6
Resources:	[PDF] from acm.org

:view less

The Second International Workshop on Open Web Search (WOWS)

2025

Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra…
European Conference on Information Retrieval, 358-365
:read more

Authors:	Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra, Martin Potthast, Arjen P de Vries, Saber Zerhoudi
Publication date:	04/03/2025
Journal:	European Conference on Information Retrieval
Pages:	358-365
Publisher:	Springer Nature Switzerland
Description:	We organize the second International Workshop on Open Web Search (WOWS) at ECIR 2025 with two calls for contributions: The first call targets scientific contributions on collaborative search engine building, including crawling, deployment, evaluation, and use of the web as a resource by researchers and innovators. The second call introduces the shared task called WOWS-Eval, which aims to gain practical experience with collaborative, cooperative search engine evaluation by focusing on enriching the Open Web Index (OWI) with relevance judgments cooperatively transferred from existing TREC-like test collections.
Cited by:	2
Resources:	[PDF] from djoerdhiemstra.com

:view less

MemBERT: Foundation model for memory forensics

2025

Christofer Fellicious, Mehdi Ben Amor, Johannes Garstenauer, Jelena Mitrović, Hans P Re…
Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing, 1772-1779
:read more

Authors:	Christofer Fellicious, Mehdi Ben Amor, Johannes Garstenauer, Jelena Mitrović, Hans P Reiser, Michael Granitzer
Publication date:	03/31/2025
Journal:	Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing
Pages:	1772-1779
Description:	Foundation models have demonstrated significant advancements in natural language processing and computer vision, yet their potential in cybersecurity is unexplored. Current memory forensics tools and machine learning models often need more versatility and adaptability, presenting a crucial research gap. To address this, we introduce MemBERT, a foundation model designed explicitly for memory forensics. MemBERT is trained on extensive process dump data, with and without metadata inclusion, to capture intricate patterns present in main memory. Its potential impact on cybersecurity practices could be significantly similar to the effects of foundation models in natural language processing. We aim to streamline memory forensics by reducing the manual effort and coding traditionally required by cybersecurity practitioners. Through comprehensive experimentation, we demonstrate MemBERT's efficiency in a …
Cited by:	1
Resources:	[PDF] from acm.org

:view less

Mixture of modular experts: Distilling knowledge from a multilingual teacher into specialized modular language models

2025

Mohammed Al-Maamari, Mehdi Ben Amor, Jelena Mitrović, Michael Granitzer
Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing, 945-852
:read more

Authors:	Mohammed Al-Maamari, Mehdi Ben Amor, Jelena Mitrović, Michael Granitzer
Publication date:	03/31/2025
Journal:	Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing
Pages:	945-852
Description:	Mixture of Experts (MoE) architectures have been increasingly employed to enhance the inference speed of Large Language Models (LLMs). However, in previous MoE approaches, there is no clear separation between the specialties of each expert, as they divide tasks based on patterns in the data rather than on specific topics. This limits their ability to specialize in particular languages or domains. Incorporating modularity by developing domain-specific experts within an MoE framework could enable more flexible and efficient LLMs, where modules can be selected and combined based on specific application needs, similar to software libraries. We present a Modular Mixture of Experts (MMoE) architecture that enhances efficiency and specialization in language models by integrating Knowledge Distillation (KD) with MoE to create domain-specific experts. In our approach, LLMs are compressed into smaller …
Cited by:	4
Resources:	[PDF] from acm.org

:view less

Searchlab: Exploring conversational and traditional search interfaces in information retrieval

2025

Saber Zerhoudi, Michael Granitzer
Proceedings of the 2025 ACM SIGIR Conference on Human Information Interaction and Retri…
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	03/24/2025
Journal:	Proceedings of the 2025 ACM SIGIR Conference on Human Information Interaction and Retrieval
Pages:	382-389
Description:	Large Language Models (LLMs) have increased the popularity of conversational search systems and traditional search engines (SERPs), making the study of user behavior and search queries across various search systems a critical research area in information retrieval (IR). However, researchers often face challenges in developing experimental search engine systems to collect user data. This paper introduces SearchLab, a modular and extensible search framework for rapid prototyping and conducting controlled user studies. SearchLab collects comprehensive user interaction data and integrates components of modern search systems with LLM capabilities, enabling seamless switching between different search paradigms while maintaining experimental reproducibility. We conducted user studies to demonstrate the framework’s capabilities, providing preliminary insights into how users adapt their search …
Cited by:	5
Resources:	[PDF] from acm.org

:view less

Instruct-to-SPARQL: A text-to-SPARQL dataset for training SPARQL Agents

2025

Mehdi Ben Amor, Alexis Strappazzon, Michael Granitzer, Elöd Egyed-Zsigmond, Jelena Mitr…
Proceedings of the 2025 ACM SIGIR Conference on Human Information Interaction and Retri…
:read more

Authors:	Mehdi Ben Amor, Alexis Strappazzon, Michael Granitzer, Elöd Egyed-Zsigmond, Jelena Mitrović
Publication date:	03/24/2025
Journal:	Proceedings of the 2025 ACM SIGIR Conference on Human Information Interaction and Retrieval
Pages:	390-395
Description:	The rapid adoption of Large Language Models (LLMs) for search engines and fact-checking platforms necessitates enhancing their output accuracy. Retrieval Augmented Generation (RAG) mitigates hallucinations but requires semantically rich repositories like Wikidata. However, there is a lack of high-quality data to fine-tune LLMs for querying such knowledge bases. To address this gap, we propose a curated dataset with 2,771 unique queries for fine-tuning LLMs to generate accurate and syntactically valid SPARQL queries from natural language instructions. This dataset, customized for interaction with Wikidata, also serves as a robust benchmark for text-to-SPARQL task evaluation. Key findings show that models generally perform better on queries with lower complexity.
Cited by:	10
Resources:	[PDF] from acm.org

:view less

Bridging the Semantic Gap in Virtual Machine Introspection and Forensic Memory Analysis

2025

Christofer Fellicious, Hans P Reiser, Michael Granitzer
arXiv preprint arXiv:2503.05482
:read more

Authors:	Christofer Fellicious, Hans P Reiser, Michael Granitzer
Publication date:	03/07/2025
Journal:	arXiv preprint arXiv:2503.05482
Description:	Forensic Memory Analysis (FMA) and Virtual Machine Introspection (VMI) are critical tools for security in a virtualization-based approach. VMI and FMA involves using digital forensic methods to extract information from the system to identify and explain security incidents. A key challenge in both FMA and VMI is the "Semantic Gap", which is the difficulty of interpreting raw memory data without specialized tools and expertise. In this work, we investigate how a priori knowledge, metadata and engineered features can aid VMI and FMA, leveraging machine learning to automate information extraction and reduce the workload of forensic investigators. We choose OpenSSH as our use case to test different methods to extract high level structures. We also test our method on complete physical memory dumps to showcase the effectiveness of the engineered features. Our features range from basic statistical features to advanced graph-based representations using malloc headers and pointer translations. The training and testing are carried out on public datasets that we compare against already recognized baseline methods. We show that using metadata, we can improve the performance of the algorithm when there is very little training data and also quantify how having more data results in better generalization performance. The final contribution is an open dataset of physical memory dumps, totalling more than 1 TB of different memory state, software environments, main memory capacities and operating system versions. Our methods show that having more metadata boosts performance with all methods obtaining an F1-Score of over 80%. Our research …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

Malware Detection based on API calls

2025

Christofer Fellicious, Manuel Bischof, Kevin Mayer, Dorian Eikenberg, Stefan Hausotte, …
arXiv preprint arXiv:2502.12863
:read more

Authors:	Christofer Fellicious, Manuel Bischof, Kevin Mayer, Dorian Eikenberg, Stefan Hausotte, Hans P Reiser, Michael Granitzer
Publication date:	02/18/2025
Journal:	arXiv preprint arXiv:2502.12863
Description:	Malware attacks pose a significant threat in today's interconnected digital landscape, causing billions of dollars in damages. Detecting and identifying families as early as possible provides an edge in protecting against such malware. We explore a lightweight, order-invariant approach to detecting and mitigating malware threats: analyzing API calls without regard to their sequence. We publish a public dataset of over three hundred thousand samples and their function call parameters for this task, annotated with labels indicating benign or malicious activity. The complete dataset is above 550GB uncompressed in size. We leverage machine learning algorithms, such as random forests, and conduct behavioral analysis by examining patterns and anomalies in API call sequences. By investigating how the function calls occur regardless of their order, we can identify discriminating features that can help us identify malware early on. The models we've developed are not only effective but also efficient. They are lightweight and can run on any machine with minimal performance overhead, while still achieving an impressive F1-Score of over 85%. We also empirically show that we only need a subset of the function call sequence, specifically calls to the ntdll.dll library, to identify malware. Our research demonstrates the efficacy of this approach through empirical evaluations, underscoring its accuracy and scalability. The code is open source and available at Github along with the dataset on Zenodo.
Cited by:	11
Resources:	[PDF] from arxiv.org

:view less

Investigating the Robustness of Embedding Models on Noisy Input Texts.

2025

Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrovic, Michael Gra…
WOWS, 2-12
:read more

Authors:	Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrovic, Michael Granitzer
Publication date:	01/01/2025
Journal:	WOWS
Pages:	2-12
Description:	Integrating dense retrieval into Retrieval-Augmented Generation systems has significantly enhanced the power of generative language models, especially in the context of conversational search. Despite this advancement, the effect of noisy and poorly structured input text, often drawn from web pages covering multiple topics or languages, has not been thoroughly investigated. This paper addresses that gap by evaluating the robustness of embedding models in handling documents under realistic, noisy conditions beyond the clean and concise golden passage found in most academic datasets. We therefore introduce a synthetic dataset based on the MS-MARCO Passage Collection, in which the original passages are concatenated with text snippets extracted from extraneous web pages. This dataset consists of multiple subcorpora, whose concatenated context varies in length, positioning and relevance to the original query-relevant passage. Our findings reveal two critical limitations to model robustness: first, a positional bias in processing long documents, and second, a significant drop in retrieval performance when models are faced with semantically inconsistent input documents that combine query-relevant information with thematically unrelated (Out-of-Domain) text passages. Motivated by our observations, we evaluate various chunking methods regarding their ability to preserve contextual and semantic integrity during index creation but find them falling short of the Naive approach in several instances. The paper calls for further research into advanced chunking techniques and alternative retrieval strategies, to improve the robustness of …
Resources:	[PDF] from ceur-ws.org

:view less

Combining Embedding Models for RAG: A Similarity-Based Approach.

2025

Kanishka Ghosh Dastidar, Michael Dinzinger, Laura Caspari, Jelena Mitrovic, Michael Gra…
WOWS, 36-49
:read more

Authors:	Kanishka Ghosh Dastidar, Michael Dinzinger, Laura Caspari, Jelena Mitrovic, Michael Granitzer
Publication date:	01/01/2025
Journal:	WOWS
Pages:	36-49
Description:	This paper explores the effective combination of embedding models through a similarity analysis. Embedding models are characterized by heterogeneous architectures, training objectives, and training corpora, resulting in distinct semantic representations and domain-specific proficiencies. By integrating such diverse models, we aim to harness their complementary strengths. However, the potential for performance improvement is contingent on the specific models selected for combination. To address this, we propose a strategy that utilizes similarity scores between model pairs as strong predictors of their combined performance. First, we provide a similarity analysis of embedding models as an unsupervised evaluation framework from a retrieval perspective. Second, we empirically demonstrate that combining embeddings from different models can significantly enhance retrieval performance, but only in the case of certain model pairs. Third, we propose an adjusted dissimilarity measure that accounts for both the similarity and performance gap between the models. Our experiments reveal that the adjusted dissimilarity scores between models are strongly correlated to their combined performance. This finding paves the way for organizations with resource constraints to use such a similarity framework to select combinations of smaller models whose performance closes the gap to large, inaccessible models.
Resources:	[PDF] from ceur-ws.org

:view less

Replication Data for: A Multimodal GeoAI Approach to Combining Text with Spatiotemporal Features for Enhanced Relevance Classification of Social Media Posts in Disaster Response

2025

David Hanny, Sebastian Schmidt, Shaily Gandhi, Michael Granitzer, Bernd Resch
Harvard Dataverse dataset, 403
:read more

Authors:	David Hanny, Sebastian Schmidt, Shaily Gandhi, Michael Granitzer, Bernd Resch
Publication date:	01/01/2025
Journal:	Harvard Dataverse dataset
Pages:	403
Description:	Annotated replication data (train/test) for our study on multimodal relevance classification of geo-social media for disaster response. Train size: 3,659 Test size: 915.

:view less

Studying memorization of large language models using answers to stack overflow questions

2025

Laura Caspari, Alexander Trautsch, Michael Granitzer, Steffen Herbold
Transactions on Machine Learning Research
:read more

Authors:	Laura Caspari, Alexander Trautsch, Michael Granitzer, Steffen Herbold
Publication date:	01/01/2025
Journal:	Transactions on Machine Learning Research
Description:	Large Language Models (LLMs) are capable of answering many software related questions and supporting developers by generating code snippets. These capabilities originate from training on massive amounts of data from the Internet, including information from Stack Overflow. This raises the question whether answers to software related questions are simply memorized from the training data, which might raise problems as this often requires attribution (e.g., CC-BY license), sharing with a similar license (e.g., GPL licenses) or may even be prohibited (proprietary license). To study this, we compare responses to questions from Stack Overflow for questions that were known during LLM pre-training and questions that were not included in the pre-training data. We then calculate the overlap both with answers marked as accepted on Stack Overflow as well as other texts we can find on the internet. We further explore the impact of the popularity of programming languages, the complexity of the prompts used, and the randomization of the text generation process on the memorization of answers to Stack Overflow. We find that many generated answers are to some degree collages of memorized content and that this does not dependent on whether the questions were seen during training or not. However, many of the memorized snippets are common phrases or code and, therefore, not copyrightable. Still, we also have clear evidence that copyright violation happens and is likely when LLMs are used at large scales.
Cited by:	3
Resources:	[PDF] from openreview.net

:view less

OpenWebSearch. eu-Building an Open Web Index on EuroHPC JU Infrastructures

2025

Michael Granitzer, Mohamad Hayek, Sebastian Heineking, Gijs Hendriksen, Martin Golasows…
Procedia Computer Science 255, 43-52
:read more

Authors:	Michael Granitzer, Mohamad Hayek, Sebastian Heineking, Gijs Hendriksen, Martin Golasowski, Michael Dinzinger, Saber Zerhoudi
Publication date:	01/01/2025
Journal:	Procedia Computer Science
Volume:	255
Pages:	43-52
Publisher:	Elsevier
Description:	The OpenWebSearch.eu project aims to develop an Open Web Index (OWI), an openly accessible data structure that supports the creation of web search engines. Building such an index requires a data- and compute-intensive pipeline for cleaning, preprocessing, enriching and indexing large amounts of web data. Beyond search, the availability of clean and preprocessed web data is also crucial for fields like web analytics and generative AI. This paper presents our approach to constructing the OWI using High-Performance Computing (HPC) resources from both EuroHPC JU and non-EuroHPC JU data centers. We contribute in two main areas: first, by detailing the development of pre-processing and indexing pipelines embedded in HPC workflows; and second, by describing the iRODS-based federated storage infrastructure and the LEXIS1 platform that will manage the cross-data centre workflows and facilitate …
Cited by:	3
Resources:	[PDF] from sciencedirect.com Find it @ Konkuk Glocal

:view less

Enhancing rhetorical figure annotation: An ontology-based web application with RAG integration

2025

Ramona Kühn, Jelena Mitrović, Michael Granitzer
Proceedings of the 31st International Conference on Computational Linguistics, 8774-8786
:read more

Authors:	Ramona Kühn, Jelena Mitrović, Michael Granitzer
Publication date:	01/01/2025
Journal:	Proceedings of the 31st International Conference on Computational Linguistics
Pages:	8774-8786
Description:	Rhetorical figures play an important role in our communication. They are used to convey subtle, implicit meaning, or to emphasize statements. We notice them in hate speech, fake news, and propaganda. By improving the systems for computational detection of rhetorical figures, we can also improve tasks such as hate speech and fake news detection, sentiment analysis, opinion mining, or argument mining. Unfortunately, there is a lack of annotated data, as well as qualified annotators that would help us build large corpora to train machine learning models for the detection of rhetorical figures. The situation is particularly difficult in languages other than English, and for rhetorical figures other than metaphor, sarcasm, and irony. To overcome this issue, we develop a web application called “Find your Figure” that facilitates the identification and annotation of German rhetorical figures. The application is based on the German Rhetorical ontology GRhOOT which we have specially adapted for this purpose. In addition, we improve the user experience with Retrieval Augmented Generation (RAG). In this paper, we present the restructuring of the ontology, the development of the web application, and the built-in RAG pipeline. We also identify the optimal RAG settings for our application. Our approach is one of the first to practically use rhetorical ontologies in combination with RAG and shows promising results.
Cited by:	3
Resources:	[PDF] from aclanthology.org

:view less

Cognitive-aware user search behavior simulation

2024

Saber Zerhoudi, Michael Granitzer
Proceedings of the 24th ACM/IEEE Joint Conference on Digital Libraries, 1-12
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	12/16/2024
Journal:	Proceedings of the 24th ACM/IEEE Joint Conference on Digital Libraries
Pages:	1-12
Description:	In Information Retrieval (IR), evolving models have begun to address the complexity of user search behavior, as traditional approaches often oversimplify the nuanced needs of web search users. While effective for simpler informational needs, these models struggle to capture the dynamic intricacies involved in complex search tasks. To address these limitations, we introduce the Cognitive-Aware Complex Searcher Model (CACSM), an advanced model that simulates user actions and integrates a deep understanding of users' cognitive states into the search process. CACSM enhances the existing Complex Searcher Model by incorporating cognitive state updates that dynamically reflect users' knowledge and information needs throughout the search session. By utilizing traditional RNN-based (CACSMum) and Large Language Model-based (CACSMllm) user model strategies, CACSM simulates user actions and …
Cited by:	7
Resources:	[PDF] from acm.org

:view less

Towards an Improved Metric for Evaluating Disentangled Representations

2024

Sahib Julka, Yashu Wang, Michael Granitzer
2024 IEEE International Conference on Knowledge Graph (ICKG), 150-158
:read more

Authors:	Sahib Julka, Yashu Wang, Michael Granitzer
Publication date:	12/11/2024
Journal:	2024 IEEE International Conference on Knowledge Graph (ICKG)
Pages:	150-158
Publisher:	IEEE
Description:	Disentangled representation learning plays a pivotal role in making representations controllable, interpretable and transferable. Despite its significance in the domain, the quest for reliable and consistent quantitative disentanglement metric remains a major challenge. This stems from the utilisation of diverse metrics measuring different properties and the potential bias introduced by their design. Our work undertakes a comprehensive examination of existing popular disentanglement evaluation metrics, comparing them in terms of measuring aspects of disentanglement (viz. Modularity, Compactness, and Explicitness), detecting the factor-code relationship, and describing the degree of disentanglement. We propose a new framework for quantifying disentanglement, introducing a metric entitled EDI, that leverages the intuitive concept of exclusivity and improved factor-code relationship to minimize ad-hoc decisions …
Resources:	[PDF] from arxiv.org

:view less

Generative Agents Navigating Digital Libraries

2024

Saber Zerhoudi, Michael Granitzer
International Conference on Asian Digital Libraries, 171-188
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	12/04/2024
Journal:	International Conference on Asian Digital Libraries
Pages:	171-188
Publisher:	Springer Nature Singapore
Description:	In the rapidly evolving field of digital libraries, the development of large language models (LLMs) has opened up new possibilities for simulating user behavior. This innovation addresses the longstanding challenge in digital library research: the scarcity of publicly available datasets on user search patterns due to privacy concerns. In this context, we introduce Agent4DL, a user search behavior simulator specifically designed for digital library environments. Agent4DL generates realistic user profiles and dynamic search sessions that closely mimic actual search strategies, including querying, clicking, and stopping behaviors tailored to specific user profiles. Our simulator’s accuracy in replicating real user interactions has been validated through comparisons with real user data. Notably, Agent4DL demonstrates competitive performance compared to existing user search simulators such as SimIIR 2.0, particularly in its …
Cited by:	4
Resources:	[PDF] from arxiv.org

:view less

Krony-PT: GPT2 compressed with Kronecker Products

2024

Mohamed Ayoub Ben Ayad, Jelena Mitrovic, Michael Granitzer
arXiv e-prints, arXiv: 2412.12351
:read more

Authors:	Mohamed Ayoub Ben Ayad, Jelena Mitrovic, Michael Granitzer
Publication date:	12/01/2024
Journal:	arXiv e-prints
Pages:	arXiv: 2412.12351
Description:	We introduce Krony-PT, a compression technique for GPT-2 based on Kronecker products. We specifically target the feed-forward weights of each transformer block, and systematically compress the feed-forward layer matrices to various degrees. We introduce a modified Van Loan decomposition to initialize new Kronecker factors, and also propose a new pruning-based initialization technique. Our method compresses the original 124M-parameter GPT-2 to various smaller models, ranging from 80M to 96M. Our 81M model variant outperforms DistilGPT2 on next-token prediction across all standard language modeling datasets, and shows competitive or comparable performance with significantly larger Kronecker-based compressions of GPT-2.

:view less

Suds: A strategy for unsupervised drift sampling

2024

Christofer Fellicious, Lorenz Wendlinger, Mario Lopez Gancarski, Jelena Mitrović, Micha…
2024 IEEE 36th International Conference on Tools with Artificial Intelligence (ICTAI), …
:read more

Authors:	Christofer Fellicious, Lorenz Wendlinger, Mario Lopez Gancarski, Jelena Mitrović, Michael Granitzer
Publication date:	10/28/2024
Journal:	2024 IEEE 36th International Conference on Tools with Artificial Intelligence (ICTAI)
Pages:	187-192
Publisher:	IEEE
Description:	Supervised machine learning often encounters concept drift, where the data distribution changes over time, degrading model performance. Existing drift detection methods focus on identifying these shifts but often overlook the challenge of acquiring labeled data for model retraining after a shift occurs. We present the Strategy for Drift Sampling (SUDS), a novel method that selects homogeneous samples for retraining using existing drift detection algorithms, thereby enhancing model adaptability to evolving data. SUDS seamlessly integrates with current drift detection techniques. We also introduce the Harmonized Annotated Data Accuracy Metric (HADAM), a metric that evaluates classifier performance about the quantity of annotated data required to achieve the stated performance, thereby taking into account the difficulty of acquiring labeled data. Our contributions are twofold: SUDS combines drift detection with …
Cited by:	1
Resources:	[PDF] from arxiv.org

:view less

Machine learning methods for credit card fraud detection: A survey

2024

Kanishka Ghosh Dastidar, Olivier Caelen, Michael Granitzer
IEEE Access 12, 158939-158965
:read more

Authors:	Kanishka Ghosh Dastidar, Olivier Caelen, Michael Granitzer
Publication date:	10/28/2024
Journal:	IEEE Access
Volume:	12
Pages:	158939-158965
Publisher:	IEEE
Description:	The widespread adoption of online payments has been accompanied by a significant increase in fraudulent activities, resulting in billions of dollars in financial losses. As payment providers aim to tackle this with various preventive mechanisms, fraudsters also continuously evolve their methods to remain indistinguishable from genuine actors. This necessitates sophisticated fraud detection tools to supplement these security mechanisms. As the volume of transactions taking place per day is in the millions, relying solely on human investigation is expensive and ultimately unfeasible, leading to an emergence of research into data driven or statistical methods for fraud detection. Over the last decade, this research has evolved to tackle the various particularities of the domain. These include the skewed nature of the data, the evolving user and fraud behavior, and the learning representations of the context in which a …
Cited by:	60
Resources:	[PDF] from ieee.org

:view less

Impact of Tokenization Techniques on URL Classification

2024

M Al-Maamari, M Istaiti, S Zerhoudi, M Dinzinger, M Granitzer, J Mitrovic
6th International Open Search Symposium# ossym2024, 44
:read more

Authors:	M Al-Maamari, M Istaiti, S Zerhoudi, M Dinzinger, M Granitzer, J Mitrovic
Publication date:	10/10/2024
Journal:	6th International Open Search Symposium# ossym2024
Pages:	44
Description:	Web crawling can be improved by the accurate classification of URLs to ensure relevant content is indexed and harmful content is filtered out. In this study, we examined the impact of various tokenization techniques on URL classification, a task integral to the development of intelligent web crawlers. Our investigation was conducted using a largescale dataset of over one million URLs, categorized into’Malicious’,’Benign’, and’Adult’classes, with detailed sublabels for in-depth analysis [1]. We explored a range of tokenization methods, including Byte Pair Encoding (BPE), Enhanced BPE with a GPT-4 generated keyword dictionary, punctuation-based splitting, and character-level n-grams, to assess their effect on the classification accuracy and computational efficiency [2, 3]. The results indicated that while simple tokenization methods like Char 1-gram offered rapid prediction times, they were inadequate in correctly identifying more complex’Malicious’ URLs. More sophisticated techniques such as BPE and WordPiece achieved a better balance of precision and recall for’Benign’and’Adult’content, yet they, along with other methods, struggled with the’Malicious’ category. The findings highlight the nuanced challenges of URL classification and underscore the need for advanced tokenization approaches that can compete with the nature of malicious content while maintaining computational efficiency. Future work should focus on integrating diverse tokenization strategies and enhancing semantic comprehension within the tokenization process to improve classification performance, particularly for detecting malicious content within the vast and dynamic …
Resources:	[PDF] from dlr.de

:view less

Federated data infrastructure for the open web search

2024

Noor A Fathima, C Ariyo, M Dinzinger, M Golasowski, M Hayek, G Hendriksen, M Karlsson, …
6th International Open Search Symposium# ossym2024, 17
:read more

Authors:	Noor A Fathima, C Ariyo, M Dinzinger, M Golasowski, M Hayek, G Hendriksen, M Karlsson, K Mankinen, S Moiras, J Truckenbrodt, L Vojacek, S Hachinger, J Martinovič, M Granitzer, A Wagner
Publication date:	10/09/2024
Journal:	6th International Open Search Symposium# ossym2024
Pages:	17
Description:	The World Wide Web, as an indispensable technological construct and infrastructure, catalyses innovation and underpins the foundation of our digital economy. It provides a comprehensive technological scaffold for a myriad of services, advancements, and enterprises, and is a pivotal data reservoir for researchers, creators, commercial entities, and the wider society. The initiative Open Web Search. eu [1], funded by the Horizon Europe programme [2], is designed to harness the untapped potential of the Web as a data repository and to construct a scalable and open European open
Cited by:	1
Resources:	[PDF] from dlr.de

:view less

Architecting the OpenSearch Service at CERN for OpenWebSearch. EU

2024

Noor A Fathima, M Dinzinger, M Granitzer, A Wagner
6th International Open Search Symposium# ossym2024, 1
:read more

Authors:	Noor A Fathima, M Dinzinger, M Granitzer, A Wagner
Publication date:	10/09/2024
Journal:	6th International Open Search Symposium# ossym2024
Pages:	1
Description:	The Open Web Search. eu (OWS) project [1], aims to harness the untapped potential of the Web as a data source and make it accessible by developing a publicly funded, scalable and open European Web search and analysis infrastructure [2]. This research initiative is a joint venture that involves 14 research institutions throughout Europe [3], with CERN serving as one of the key technical infrastructure partners. The mission includes a robust, highly available continuous crawling operation, adhering to fundamental principles of openness and legal compliance. To this end, the OWS team has created the Open Web Crawler (OWLer)[4], an open-source software framework that facilitates the extensive collection of Web documents [5]. Amongst other, the OWLer comprises the URL Frontier Service which has the data structure containing all URLs discovered during the crawling process, effectively acting as the project’s brain. It uses OpenSearch [6] technology as its primary back-end chosen for its robust persistence, distribution, and replication capabilities along with its indexing and querying capabilities, many of its open source plugins, and to leverage an existing OpenSearch-based implementation [7] of the open source URLFrontier framework [8]–which has been funded by NGI Zero [9]. The OpenSearch service is continuously operational at the CERN data center, and this article details along with our experience, it’s technical setup and evolution throughout different phases of the project, including the design, implementation, improvements made and the challenges we overcame. This has culminated in us effectively parsing, indexing, and storing∼ …
Cited by:	1
Resources:	[PDF] from dlr.de

:view less

A dataset of GDPR compliant NER for privacy policies

2024

Harshil Darji, Stefan Becher, Jelena Mitrović, Armin Gerl, Michael Granitzer
Proceedings of the 6th International Open Search Symposium (ossym24). CERN, Garching (M…
:read more

Authors:	Harshil Darji, Stefan Becher, Jelena Mitrović, Armin Gerl, Michael Granitzer
Publication date:	10/09/2024
Journal:	Proceedings of the 6th International Open Search Symposium (ossym24). CERN, Garching (Munich), Germany
Pages:	26-31
Description:	Privacy policies play a vital role in informing users about the data practices of online platforms. They are intended to help them make informed decisions regarding the processing of their personal information. Still, privacy policies are often long and complicated, making it difficult for users to understand how their data is being handled. Natural Language Processing (NLP) techniques, such as Named Entity Recognition (NER), can be employed to automatically extract meaningful information from privacy policies to ease the making of informed decisions. In this work, we present a dataset of privacy policies improved with NER annotations. The dataset consists of privacy policies from 44 online platforms. These policies were annotated to comply with the GDPR guidelines. The privacy policies are manually annotated with NER tags, highlighting relevant entities of GDPR privacy policies such as data controllers, data sources, authority, etc. We also provide the annotation guidelines used by the annotators. This annotated dataset is a valuable resource for training and evaluating NER models in the context of privacy policies.
Cited by:	2
Resources:	[PDF] from dlr.de

:view less

Comparative analysis: User interactions in public and private digital libraries datasets

2024

Saber Zerhoudi, Michael Granitzer
International Conference on Theory and Practice of Digital Libraries, 162-172
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	09/24/2024
Journal:	International Conference on Theory and Practice of Digital Libraries
Pages:	162-172
Publisher:	Springer Nature Switzerland
Description:	The scarcity of public datasets with detailed user interaction data in digital libraries underscores a significant research challenge. The complexity of effectively anonymizing such data without losing its utility further exacerbates this issue. This study conducts a comparative analysis of two datasets: EconBiz (https://econbiz.de), a private dataset, and SUSS (The dataset is publicly available at http://dx.doi.org/10.7802/1380) a publicly available dataset, to investigate the differences in the granularity of user logs between them. The findings reveal that while EconBiz and SUSS share some similar patterns, EconBiz provides a more nuanced view of diverse user interactions compared to SUSS. These results emphasize the need for enhancing public datasets to maintain their value for research and real-world applications. Moreover, this paper briefly explores the potential of using few-shot prompting with large language …
Cited by:	3

:view less

MCGRAN: Multi-conditional Graph Generation for Neural Architecture Search

2024

Sathish Purushothaman, Julian Stier, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 302-316
:read more

Authors:	Sathish Purushothaman, Julian Stier, Michael Granitzer
Publication date:	09/21/2024
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	302-316
Publisher:	Springer Nature Switzerland
Description:	We propose Multi-Conditional Graph Recurrent Attention Networks (MCGRAN) to learn conditional distributions over labeled graphs and show their application in Neural Architecture Search. MCGRAN learns neural network architectures as labeled directed acyclic graphs. With proper conditioning MCGRAN can generate neural network architectures with different performance characteristics. MCGRAN generates architectures with test accuracies of 0.90±0.01 in a high performing regime but also meets test accuracies of 0.53±0.17 when specifically conditioned on low performing architectures. With graph edit distances we demonstrate that 99% of the generated neural network architectures are novel and unique. Using an iterative conditional search, we show that neural network architectures can hierarchically be decomposed in terms of model parameters, training time, and accuracy. Also, we apply MCGRAN to …
Cited by:	1

:view less

Leveraging Graph Networks and Generative Adversarial Networks for Controllable Trajectory Predictions

2024

Sahib Julka, Ahmad Zubair, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 178-192
:read more

Authors:	Sahib Julka, Ahmad Zubair, Michael Granitzer
Publication date:	09/21/2024
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	178-192
Publisher:	Springer Nature Switzerland
Description:	The rapid advancement in autonomous technologies highlights the growing importance of accurate pedestrian trajectory prediction. While Generative models have demonstrated good performance in this task, their latent spaces often lack controllability. Addressing this gap, this work presents an innovative framework exploiting Generative Adversarial Networks with control vectors such as InfoGAN and augmenting them with Graph embeddings to obtain more accurate representations, which can be disentangled and improve accuracy in downstream tasks. Evaluation against benchmark datasets demonstrates good predictive accuracy for pedestrian trajectory prediction compared to current state-of-the-art approaches. Furthermore, we show that this method can enable contraollability of disentangled factors. This research lays the groundwork for developing advanced simulation tools with various real-world …

:view less

PointerKex: a pointer-based ssh key extraction method

2024

Christofer Fellicious, Stewart Sentanoe, Jelena Mitrović, Hans P Reiser, Michael Granit…
International Conference on Machine Learning, Optimization, and Data Science, 272-286
:read more

Authors:	Christofer Fellicious, Stewart Sentanoe, Jelena Mitrović, Hans P Reiser, Michael Granitzer
Publication date:	09/21/2024
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	272-286
Publisher:	Springer Nature Switzerland
Description:	Most SSH key extraction methods either use modified source code or tools like active Man-in-the-Middle proxies. Here, we present our method to extract SSH Encryption keys using features engineered from structures pointed to by pointers within the heap. The only additional information required in our method is a way to translate addresses contained by pointers present in the heap. We then create a machine-learning model to identify representative structures within the heap data allowing us direct access to the keys. Our proposed method retrieves the SSH Keys, lengths, and encryption algorithm while maintaining perfect precision and recall. Our method also has an execution time far less than comparable methods. We present our findings and studies including the minimum, maximum and average metrics on different sized subsets of the dataset.
Cited by:	2

:view less

Llms in the loop: Leveraging large language model annotations for active learning in low-resource languages

2024

Nataliia Kholodna, Sahib Julka, Mohammad Khodadadi, Muhammed Nurullah Gumus, Michael Gr…
Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 397…
:read more

Authors:	Nataliia Kholodna, Sahib Julka, Mohammad Khodadadi, Muhammed Nurullah Gumus, Michael Granitzer
Publication date:	08/22/2024
Journal:	Joint European Conference on Machine Learning and Knowledge Discovery in Databases
Pages:	397-412
Publisher:	Springer Nature Switzerland
Description:	Low-resource languages face significant barriers in AI development due to limited linguistic resources and expertise for data labeling, rendering them rare and costly. The scarcity of data and the absence of preexisting tools exacerbate these challenges, especially since these languages may not be adequately represented in various NLP datasets. To address this gap, we propose leveraging the potential of LLMs in the active learning loop for data annotation. Initially, we conduct evaluations to assess inter-annotator agreement and consistency, facilitating the selection of a suitable LLM annotator. The chosen annotator is then integrated into a training loop for a classifier using an active learning paradigm, minimizing the amount of queried data required. Empirical evaluations, notably employing GPT-4-Turbo, demonstrate near-state-of-the-art performance with significantly reduced data requirements, as indicated by …
Cited by:	76
Resources:	[PDF] from arxiv.org

:view less

Report on the 1st international workshop on open web search (WOWS 2024) at ECIR 2024

2024

Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra…
ACM SIGIR Forum 58 (1), 1-13
:read more

Authors:	Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra, Martin Potthast, Arjen P de Vries, Saber Zerhoudi
Publication date:	08/07/2024
Journal:	ACM SIGIR Forum
Volume:	58
Issue:	1
Pages:	1-13
Publisher:	ACM
Description:	The first International Workshop on Open Web Search (WOWS) was held on Thursday, March 28th, at ECIR 2024 in Glasgow, UK. The full-day workshop had two calls for contributions: the first call aimed at scientific contributions to building, operating, and evaluating search engines cooperatively and the cooperative use of the web as a resource for researchers and innovators. The second call for implementations of retrieval components aimed to gain practical experience with joint, cooperative evaluation of search engines and their components. In total, 2 papers were accepted for the first call, and 11 software components were submitted for the second. The workshop ended with breakout sessions on how the Open-WebSearch.eu project can incorporate collaborative evaluations and a hub of search engines. Date: 28 March 2024. Website: https://opensearchfoundation.org/wows2024/.
Cited by:	3
Resources:	[PDF] from acm.org

:view less

Method for intrusion detection to detect malicious insider threat activities and system for intrusion detection

2024

:read more

Publication date:	08/06/2024
Description:	(57) ABSTRACT A method and system for intrusion detection to detect malicious insider threat activities within a network of user profiles. The method includes training a Neural Network on multiple sets of user profile data for multiple user profiles and on multiple sets of activity data of the multiple user profiles of the network, such that the Neural Network is capable of predicting for future dates activities for multiple user profiles. The method includes applying the trained Neural Network on the set of further user profile data of the further user profile, predicting an activity of the further user profile based on the multiple sets of activity data by the trained Neural Network, observing activity of the further user profile, applying the trained Neural Network on the observed activity, and detecting malicious activity for the further user profile by the trained Neural Network, if the observed activity deviates from the predicted activity.
Resources:	[PDF] from googleapis.com

:view less

Personarag: Enhancing retrieval-augmented generation systems with user-centric agents

2024

Saber Zerhoudi, Michael Granitzer
arXiv preprint arXiv:2407.09394
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	07/12/2024
Journal:	arXiv preprint arXiv:2407.09394
Description:	Large Language Models (LLMs) struggle with generating reliable outputs due to outdated knowledge and hallucinations. Retrieval-Augmented Generation (RAG) models address this by enhancing LLMs with external knowledge, but often fail to personalize the retrieval process. This paper introduces PersonaRAG, a novel framework incorporating user-centric agents to adapt retrieval and generation based on real-time user data and interactions. Evaluated across various question answering datasets, PersonaRAG demonstrates superiority over baseline models, providing tailored answers to user needs. The results suggest promising directions for user-adapted information retrieval systems.
Cited by:	45
Resources:	[PDF] from arxiv.org

:view less

Can GPT-4 Replace Human Examiners? A Competition on Checking Open-Text Answers

2024

Abdullah Al Zubaer, Michael Granitzer, Stephan Geschwind, Johann Graf Lambsdorff, Debor…
:read more

Authors:	Abdullah Al Zubaer, Michael Granitzer, Stephan Geschwind, Johann Graf Lambsdorff, Deborah Voss
Publication date:	07/11/2024
Description:	Generative AI models have taken the world by storm. ChatGPT 1, for instance, surpassed the mark of 1 million users within five 12 days of its launch in December 2022 and reached 100 million users within two months. Its rapid spread can be attributed to the 13 high quality of human-like conversations. It has the potential of broad repercussions for society in general and the educational 14 system in particular. Especially the fourth version of GPT (Generative Pre-trained Transformer) 2 launched in March 2023, is 15 widely acknowledged for its ability to understand and produce text. One of GPT’s fundamental properties is its capacity to 16 perform in-context learning 3 in which it “learns” from a prompt containing examples or instructions to perform specific tasks 17 relying entirely on its linguistic capability 4. This allows GPT models, particularly GPT-4 5, to be highly adaptable in various 18 scenarios despite being a generalized language model. GPT-4, therefore, has a multitude of potential use cases in the education 19 system. It can be used both by students and teachers to assist or even replace them in writing and grading exams and papers. 20 Accordingly, studies on the usage of generative AI models in education have mushroomed after the release of ChatGPT. 21 When it comes to using GPT as a student, Herbold et al. 6 demonstrate the capability of GPT-3.5 and GPT-4 to write 22 high-quality essays. In a large-scale project, they have more than 100 high school teachers compare the performance of essays 23 written by advanced cohorts of high school students to essays written by GPT and find that the latter surpasses humans in terms 24 of …
Cited by:	2
Resources:	[PDF] from researchsquare.com

:view less

Beyond benchmarks: Evaluating embedding model similarity for retrieval augmented generation systems

2024

Laura Caspari, Kanishka Ghosh Dastidar, Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
arXiv preprint arXiv:2407.08275
:read more

Authors:	Laura Caspari, Kanishka Ghosh Dastidar, Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
Publication date:	07/11/2024
Journal:	arXiv preprint arXiv:2407.08275
Description:	The choice of embedding model is a crucial step in the design of Retrieval Augmented Generation (RAG) systems. Given the sheer volume of available options, identifying clusters of similar models streamlines this model selection process. Relying solely on benchmark performance scores only allows for a weak assessment of model similarity. Thus, in this study, we evaluate the similarity of embedding models within the context of RAG systems. Our assessment is two-fold: We use Centered Kernel Alignment to compare embeddings on a pair-wise level. Additionally, as it is especially pertinent to RAG systems, we evaluate the similarity of retrieval results between these models using Jaccard and rank similarity. We compare different families of embedding models, including proprietary ones, across five datasets from the popular Benchmark Information Retrieval (BEIR). Through our experiments we identify clusters of models corresponding to model families, but interestingly, also some inter-family clusters. Furthermore, our analysis of top-k retrieval similarity reveals high-variance at low k values. We also identify possible open-source alternatives to proprietary models, with Mistral exhibiting the highest similarity to OpenAI models.
Cited by:	49
Resources:	[PDF] from arxiv.org

:view less

Driftgan: Using historical data for unsupervised recurring drift detection

2024

Christofer Fellicious, Sahib Julka, Lorenz Wendlinger, Michael Granitzer
arXiv preprint arXiv:2407.06543
:read more

Authors:	Christofer Fellicious, Sahib Julka, Lorenz Wendlinger, Michael Granitzer
Publication date:	07/09/2024
Journal:	arXiv preprint arXiv:2407.06543
Description:	In real-world applications, input data distributions are rarely static over a period of time, a phenomenon known as concept drift. Such concept drifts degrade the model's prediction performance, and therefore we require methods to overcome these issues. The initial step is to identify concept drifts and have a training method in place to recover the model's performance. Most concept drift detection methods work on detecting concept drifts and signalling the requirement to retrain the model. However, in real-world cases, there could be concept drifts that recur over a period of time. In this paper, we present an unsupervised method based on Generative Adversarial Networks(GAN) to detect concept drifts and identify whether a specific concept drift occurred in the past. Our method reduces the time and data the model requires to get up to speed for recurring drifts. Our key results indicate that our proposed model can outperform the current state-of-the-art models in most datasets. We also test our method on a real-world use case from astrophysics, where we detect the bow shock and magnetopause crossings with better results than the existing methods in the domain.
Cited by:	7
Resources:	[PDF] from arxiv.org

:view less

Challenges and considerations in annotating legal data: A comprehensive overview

2024

Harshil Darji, Jelena Mitrović, Michael Granitzer
arXiv preprint arXiv:2407.17503
:read more

Authors:	Harshil Darji, Jelena Mitrović, Michael Granitzer
Publication date:	07/05/2024
Journal:	arXiv preprint arXiv:2407.17503
Description:	The process of annotating data within the legal sector is filled with distinct challenges that differ from other fields, primarily due to the inherent complexities of legal language and documentation. The initial task usually involves selecting an appropriate raw dataset that captures the intricate aspects of legal texts. Following this, extracting text becomes a complicated task, as legal documents often have complex structures, footnotes, references, and unique terminology. The importance of data cleaning is magnified in this context, ensuring that redundant information is eliminated while maintaining crucial legal details and context. Creating comprehensive yet straightforward annotation guidelines is imperative, as these guidelines serve as the road map for maintaining uniformity and addressing the subtle nuances of legal terminology. Another critical aspect is the involvement of legal professionals in the annotation process. Their expertise is valuable in ensuring that the data not only remains contextually accurate but also adheres to prevailing legal standards and interpretations. This paper provides an expanded view of these challenges and aims to offer a foundational understanding and guidance for researchers and professionals engaged in legal data annotation projects. In addition, we provide links to our created and fine-tuned datasets and language models. These resources are outcomes of our discussed projects and solutions to challenges faced while working on them.
Cited by:	11
Resources:	[PDF] from arxiv.org

:view less

A longitudinal study of content control mechanisms

2024

Michael Dinzinger, Michael Granitzer
Companion Proceedings of the ACM Web Conference 2024, 1382-1387
:read more

Authors:	Michael Dinzinger, Michael Granitzer
Publication date:	05/13/2024
Journal:	Companion Proceedings of the ACM Web Conference 2024
Pages:	1382-1387
Description:	As generative AI continues to evolve, it becomes increasingly important for site owners to effectively communicate their conditions and preferences to web agents to maintain data sovereignty. This necessity underscores the importance of an ecosystem where the technical means to prevent unauthorized data mining and to set conditions on the usage of web resources are readily available. Our research focuses on the temporal development of such technical content control methods, examining two primary mechanisms: the regulation of web robots via the Robots Exclusion Protocol and the semantic annotation of web documents with licensing information. Through a longitudinal study, we analyze the implementation and recent modifications of robots.txt files, robot directives (such as noindex, nofollow, etc.), and license-related HTML annotations. This study is driven by the growing awareness among site owners …
Cited by:	11
Resources:	[PDF] from acm.org

:view less

Using pre-trained language models in an end-to-end pipeline for antithesis detection

2024

Ramona Kühn, Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Proceedings of the 2024 Joint International Conference on Computational Linguistics, La…
:read more

Authors:	Ramona Kühn, Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Publication date:	05/01/2024
Journal:	Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
Pages:	17310-17320
Description:	Rhetorical figures play an important role in influencing readers and listeners. Some of these word constructs that deviate from the usual language structure are known to be persuasive–antithesis is one of them. This figure combines parallel phrases with opposite ideas or words to highlight a contradiction. By identifying this figure, persuasive actors can be better identified. For this task, we create an annotated German dataset for antithesis detection. The dataset consists of posts from a Telegram channel criticizing the COVID-19 politics in Germany. Furthermore, we propose a three-block pipeline approach to detect the figure antithesis using large language models. Our pipeline splits the text into phrases, identifies phrases with a syntactically parallel structure, and detects if these parallel phrase pairs present opposing ideas by fine-tuning the German ELECTRA model, a state-of-the-art deep learning model for the German language. Furthermore, we compare the results with multilingual BERT and German BERT. Our novel approach outperforms the state-of-the-art methods (F1-score of 50.43%) for antithesis detection by achieving an F1-score of 65.11%.
Cited by:	4
Resources:	[PDF] from aclanthology.org

:view less

Impact and development of an open web index for open web search

2024

Michael Granitzer, Stefan Voigt, Noor Afshan Fathima, Martin Golasowski, Christian Guet…
Journal of the Association for Information Science and Technology 75 (5), 512-520
:read more

Authors:	Michael Granitzer, Stefan Voigt, Noor Afshan Fathima, Martin Golasowski, Christian Guetl, Tobias Hecking, Gijs Hendriksen, Djoerd Hiemstra, Jan Martinovič, Jelena Mitrović, Izidor Mlakar, Stavros Moiras, Alexander Nussbaumer, Per Öster, Martin Potthast, Marjana Senčar Srdič, Sharikadze Megi, Kateřina Slaninová, Benno Stein, Arjen P de Vries, Vít Vondrák, Andreas Wagner, Saber Zerhoudi
Publication date:	05/01/2024
Journal:	Journal of the Association for Information Science and Technology
Volume:	75
Issue:	5
Pages:	512-520
Publisher:	John Wiley & Sons, Inc.
Description:	Web search is a crucial technology for the digital economy. Dominated by a few gatekeepers focused on commercial success, however, web publishers have to optimize their content for these gatekeepers, resulting in a closed ecosystem of search engines as well as the risk of publishers sacrificing quality. To encourage an open search ecosystem and offer users genuine choice among alternative search engines, we propose the development of an Open Web Index (OWI). We outline six core principles for developing and maintaining an open index, based on open data principles, legal compliance, and collaborative technology development. The combination of an open index with what we call declarative search engines will facilitate the development of vertical search engines and innovative web data products (including, e.g., large language models), enabling a fair and open information space. This framework …
Cited by:	52
Resources:	[PDF] from wiley.com

:view less

Efficient NAS with FaDE on hierarchical spaces

2024

Simon Neumeyer, Julian Stier, Michael Granitzer
International symposium on intelligent data analysis, 158-170
:read more

Authors:	Simon Neumeyer, Julian Stier, Michael Granitzer
Publication date:	04/16/2024
Journal:	International symposium on intelligent data analysis
Pages:	158-170
Publisher:	Springer Nature Switzerland
Description:	Neural architecture search (NAS) is a challenging problem. Hierarchical search spaces allow for cheap evaluations of neural network sub modules to serve as surrogate for architecture evaluations. Yet, sometimes the hierarchy is too restrictive or the surrogate fails to generalize. We present FaDE which uses differentiable architecture search to obtain relative performance predictions on finite regions of a hierarchical NAS space. The relative nature of these ranks calls for a memory-less, batch-wise outer search algorithm for which we use an evolutionary algorithm with pseudo-gradient descent. FaDE is especially suited on deep hierarchical, respectively multi-cell search spaces, which it can explore by linear instead of exponential cost and therefore eliminates the need for a proxy search space. Our experiments show that firstly, FaDE-ranks on finite regions of the search space correlate with corresponding …
Cited by:	2
Resources:	[PDF] from arxiv.org

:view less

Informed heterogeneous attention networks for metapath based learning

2024

Lorenz Wendlinger, Michael Granitzer
Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing, 458-465
:read more

Authors:	Lorenz Wendlinger, Michael Granitzer
Publication date:	04/08/2024
Journal:	Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing
Pages:	458-465
Description:	Metapath based processing is popular for heterogeneous graphs as it can be focused on relevant relations. However, recent work has shown that important information about the intermediate nodes that form a metapath is lost in the process. This puts it at a disadvantage compared to homogeneous graph processing, where all node information is available. We propose a novel attention mechanism that can be used to incorporate structural intermediate node information into metapath based attention. Combined with a more efficient propagation and aggregation strategy, this improves the performance of metapath based processing on heterogeneous graphs. These adaptations allow us to surpass state-of-the-art performance in node classification tasks from the heterogeneous graph bench-marking suite by up to 2%. We further improve upon the original Heterogeneous Attention Network by up to 8%. The used …
Cited by:	1
Resources:	[PDF] from acm.org

:view less

Impact of position bias on language models in token classification

2024

Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović
Proceedings of the 39th ACM/SIGAPP symposium on applied computing, 741-745
:read more

Authors:	Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović
Publication date:	04/08/2024
Journal:	Proceedings of the 39th ACM/SIGAPP symposium on applied computing
Pages:	741-745
Description:	Language Models (LMs) have shown state-of-the-art performance in Natural Language Processing (NLP) downstream tasks such as Named Entity Recognition (NER) or Part-of-Speech (POS) tagging. Those tasks are known to suffer from data imbalance issues, particularly regarding the ratio of positive to negative examples and class disparities. This paper investigates an often-overlooked issue of encoder models, specifically the position bias of positive examples in token classification. We propose an evaluation approach to investigate position bias in transformer models with different position embedding techniques. We show that LMs can suffer from this bias with an average drop in performance ranging from 3% to 5%. We propose two methods: Random Position Shifting and Context Perturbation, that we apply on batches during the training process. The results show an improvement of ≈ 2% in the performance …
Cited by:	7
Resources:	[PDF] from acm.org

:view less

A survey of web content control for generative AI

2024

Michael Dinzinger, Florian Heß, Michael Granitzer
arXiv preprint arXiv:2404.02309
:read more

Authors:	Michael Dinzinger, Florian Heß, Michael Granitzer
Publication date:	04/02/2024
Journal:	arXiv preprint arXiv:2404.02309
Description:	The groundbreaking advancements around generative AI have recently caused a wave of concern culminating in a row of lawsuits, including high-profile actions against Stability AI and OpenAI. This situation of legal uncertainty has sparked a broad discussion on the rights of content creators and publishers to protect their intellectual property on the web. European as well as US law already provides rough guidelines, setting a direction for technical solutions to regulate web data use. In this course, researchers and practitioners have worked on numerous web standards and opt-out formats that empower publishers to keep their data out of the development of generative AI models. The emerging AI/ML opt-out protocols are valuable in regards to data sovereignty, but again, it creates an adverse situation for a site owners who are overwhelmed by the multitude of recent ad hoc standards to consider. In our work, we want to survey the different proposals, ideas and initiatives, and provide a comprehensive legal and technical background in the context of the current discussion on web publishers control.
Cited by:	9
Resources:	[PDF] from arxiv.org

:view less

The open web index: Crawling and indexing the web for public use

2024

Gijs Hendriksen, Michael Dinzinger, Sheikh Mastura Farzana, Noor Afshan Fathima, Maik F…
European Conference on Information Retrieval, 130-143
:read more

Authors:	Gijs Hendriksen, Michael Dinzinger, Sheikh Mastura Farzana, Noor Afshan Fathima, Maik Fröbe, Sebastian Schmidt, Saber Zerhoudi, Michael Granitzer, Matthias Hagen, Djoerd Hiemstra, Martin Potthast, Benno Stein
Publication date:	03/23/2024
Journal:	European Conference on Information Retrieval
Pages:	130-143
Publisher:	Springer Nature Switzerland
Description:	Only few search engines index the Web at scale. Third parties who want to develop downstream applications based on web search fully depend on the terms and conditions of the few vendors. The public availability of the large-scale Common Crawl does not alleviate the situation, as it is often cheaper to crawl and index only a smaller collection focused on a downstream application scenario than to build and maintain an index for a general collection the size of the Common Crawl. Our goal is to improve this situation by developing the Open Web Index. The Open Web Index is a publicly funded basic infrastructure from which downstream applications will be able to select and compile custom indexes in a simple and transparent way. Our goal is to establish the Open Web Index along with associated data products as a new open web information intermediary. In this paper, we present our first prototype for the Open …
Cited by:	21
Resources:	[PDF] from dlr.de

:view less

The first international workshop on open web search (WOWS)

2024

Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra…
European Conference on Information Retrieval, 426-431
:read more

Authors:	Sheikh Mastura Farzana, Maik Fröbe, Michael Granitzer, Gijs Hendriksen, Djoerd Hiemstra, Martin Potthast, Saber Zerhoudi
Publication date:	03/23/2024
Journal:	European Conference on Information Retrieval
Pages:	426-431
Publisher:	Springer Nature Switzerland
Description:	We organize the first international Workshop on Open Web Search (WOWS) at ECIR 2024 with two calls for contributions. The first call targets scientific contributions on cooperative search engine development.This includes cooperative crawling of the web and cooperative deployment and evaluation of search engines. We specifically highlight the potential of enabling public and commercial organizations to use an indexed web crawl as a resource to create innovative search engines tailored to specific user groups, instead of relying on one search engine provider. The second call aims at gaining practical experience with joint, cooperative evaluation of search engine prototypes and their components using the Information Retrieval Experiment Platform (TIREx).
Cited by:	7
Resources:	[PDF] from djoerdhiemstra.com

:view less

PyPads: Transparent Machine Learning Experiment Tracking

2024

Thomas Weißgerber, Mehdi Ben Amor, Christofer Fellicious, Michael Granitzer
Datenbank-Spektrum 24 (1), 53-62
:read more

Authors:	Thomas Weißgerber, Mehdi Ben Amor, Christofer Fellicious, Michael Granitzer
Publication date:	03/01/2024
Journal:	Datenbank-Spektrum
Volume:	24
Issue:	1
Pages:	53-62
Publisher:	Springer Berlin Heidelberg
Description:	Despite algorithmic advancements in the field of machine learning, a need for improvement in the infrastructure supporting machine learning development and research has become increasingly apparent. Machine learning experiments usually tend to be more ad-hoc in nature, and results are communicated most often in the form of a publication. Experimental details are often omitted due to size or time constraints, or simply because the complexity in terms of technical setup or parametrization became intractable. Even access to code bases, disregard important properties of the environment and experimental setup, like for example random generators or computing infrastructure. At the same time, tracking and communicating an often inherently exploratory scientific process is a task with considerable effort. We explored different venues to tackle these issues from a data science engineering point of view. The efforts …
Cited by:	2
Resources:	[HTML] from springer.com

:view less

Pathways of Machine Learning-Based Agents on Diverse Communication Structures

2024

Sathish Purushothaman, Michael Granitzer, Florian Lemmerich, Jelena Mitrović
International Conference on Agents and Artificial Intelligence, 109-125
:read more

Authors:	Sathish Purushothaman, Michael Granitzer, Florian Lemmerich, Jelena Mitrović
Publication date:	02/24/2024
Journal:	International Conference on Agents and Artificial Intelligence
Pages:	109-125
Publisher:	Springer Nature Switzerland
Description:	The recent popularity of multi-agent conversation frameworks enabled by large language models such as AutoGen promises distributed problem-solving. However, these frameworks shed little light on how language emerges and how efficient the communication structure of different agents should be. In this work, we study communication among agents under various communication structures, where agents are transformer-based auto-encoders. Specifically, we study the emergence of language in stars, rings, wheels, randoms, and fully-connected structures. In each communication structure, a sender agent observes an image and sends a sequence of symbols to the receiver agent. The receiver agent reconstructs the image using the symbols. We evaluate the quality of the image using an MNIST classifier, which acts as the world model. Our results show that the star structure performs best, with agents achieving …

:view less

Beyond Conventional Metrics: Assessing User Simulators in Information Retrieval.

2024

Saber Zerhoudi, Michael Granitzer
IIR, 3-12
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	01/01/2024
Journal:	IIR
Pages:	3-12
Description:	Traditional evaluation methods for user simulators in Information Retrieval systems have limitations in assessing their reliability for comparative analysis. To address this, we apply the Fréchet Distance (FD) to measure similarity between real and simulated user search session distributions. Using the TREC Session 2014 Track dataset, we compare FD’s performance against established metrics like session nDCG and Expected Global Utility. Our study explores FD’s effectiveness in various scenarios, including those with minimal and extensive interaction data, and examines its sensitivity to different feature extraction methods. Results show that FD correlates strongly with existing metrics while offering unique insights into session similarity, particularly for complex, multi-query sessions. FD demonstrates robustness across feature extraction techniques and versatility in various evaluation scenarios. This research contributes to the field of Interactive Information Retrieval (IIR) by providing a more comprehensive framework for evaluating simulated search sessions.
Cited by:	3
Resources:	[PDF] from ceur-ws.org

:view less

WOWS 2024: Workshop on Open Web Search 2024, Proceedings of the first International Workshop on Open Web Search co-located with the 46th European Conference on Information …

2024

S Mastura Farzana, M Fröbe, G Hendriksen, M Granitzer, D Hiemstra, M Potthast, S Zerhoudi
:read more

Authors:	S Mastura Farzana, M Fröbe, G Hendriksen, M Granitzer, D Hiemstra, M Potthast, S Zerhoudi
Publication date:	01/01/2024
Publisher:	Sl: CEUR
Description:	This volume contains the papers of WOWS 2024, the first International Workshop on Open Web Search, co-located with the 46th European Conference on Information Retrieval (ECIR) in Glasgow, UK, on 28 March 2024. WOWS 2024 had two calls for contributions. The first call targets scientific contributions on cooperative search engine development. This includes cooperative crawling of the web and cooperative deployment and evaluation of search engines. We specifically highlight the potential of enabling public and commercial organizations to use an indexed web crawl as a resource to create innovative search engines tailored to specific user groups, instead of relying on one search engine provider. The second call aims at gaining practical experience with joint, cooperative evaluation of search engine prototypes and their components using the Information Retrieval Experiment Platform TIREx.

:view less

A Framework for Studying Communication Pathways in Machine Learning-Based Agent-to-Agent Communication.

2024

Sathish Purushothaman, Michael Granitzer, Florian Lemmerich, Jelena Mitrovic
ICAART (1), 117-126
:read more

Authors:	Sathish Purushothaman, Michael Granitzer, Florian Lemmerich, Jelena Mitrovic
Publication date:	01/01/2024
Journal:	ICAART (1)
Pages:	117-126
Description:	The rise of Large Language Models (LLMs) has increased the relevance of agent-to-agent communication, particularly in systems where agents learn from their interactions. However, current LLMs offer limited insights into the communication dynamics among multiple agents, especially in large-scale settings when multiple agents are involved. Particularly training LLMs-in contrast to in-context learning-becomes nearly infeasible without large-scale computing infrastructure. In our work we present a machine-learning based agent framework to investigate the role of different communication pathways for studying language emergence between machine learning-based agents. We designed a transformer-based image auto-encoder as the agent architecture. A Gumbel SoftMax layer encodes images in form of symbols forming the language between agents. We study two pathways: In the first pathway, the sender reads an image and sends a message to the receiver. The receiver uses the message to reconstruct the sender’s image. In the second pathway, the sender and receiver read an image and minimize the distance between the generated symbols. In the first pathway, language emerges with the Levenshtein distance of≤ 2 for 96% of messages. In the second pathway, no language emerges with the Levenshtein distance of≤ 2 for 3% of messages.
Cited by:	1
Resources:	[PDF] from github.io

:view less

Is GPT-4 a reliable rater? Evaluating consistency in GPT-4’s text ratings

2023

Veronika Hackl, Alexandra Elena Müller, Michael Granitzer, Maximilian Sailer
Frontiers in Education 8, 1272229
:read more

Authors:	Veronika Hackl, Alexandra Elena Müller, Michael Granitzer, Maximilian Sailer
Publication date:	12/05/2023
Journal:	Frontiers in Education
Volume:	8
Pages:	1272229
Publisher:	Frontiers Media SA
Description:	This study reports the Intraclass Correlation Coefficients of feedback ratings produced by OpenAI's GPT-4, a large language model (LLM), across various iterations, time frames, and stylistic variations. The model was used to rate responses to tasks related to macroeconomics in higher education (HE), based on their content and style. Statistical analysis was performed to determine the absolute agreement and consistency of ratings in all iterations, and the correlation between the ratings in terms of content and style. The findings revealed high interrater reliability, with ICC scores ranging from 0.94 to 0.99 for different time periods, indicating that GPT-4 is capable of producing consistent ratings. The prompt used in this study is also presented and explained.
Cited by:	165
Resources:	[HTML] from frontiersin.org

:view less

Towards measuring representational similarity of large language models

2023

Max Klabunde, Mehdi Ben Amor, Michael Granitzer, Florian Lemmerich
arXiv preprint arXiv:2312.02730
:read more

Authors:	Max Klabunde, Mehdi Ben Amor, Michael Granitzer, Florian Lemmerich
Publication date:	12/05/2023
Journal:	arXiv preprint arXiv:2312.02730
Description:	Understanding the similarity of the numerous released large language models (LLMs) has many uses, e.g., simplifying model selection, detecting illegal model reuse, and advancing our understanding of what makes LLMs perform well. In this work, we measure the similarity of representations of a set of LLMs with 7B parameters. Our results suggest that some LLMs are substantially different from others. We identify challenges of using representational similarity measures that suggest the need of careful study of similarity scores to avoid false conclusions.
Cited by:	19
Resources:	[PDF] from arxiv.org

:view less

Learn from one specialized sub-teacher: One-to-one mapping for feature-based knowledge distillation

2023

Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Findings of the Association for Computational Linguistics: EMNLP 2023, 13235-13245
:read more

Authors:	Khouloud Saadi, Jelena Mitrović, Michael Granitzer
Publication date:	12/01/2023
Journal:	Findings of the Association for Computational Linguistics: EMNLP 2023
Pages:	13235-13245
Description:	Knowledge distillation is known as an effective technique for compressing over-parameterized language models. In this work, we propose to break down the global feature distillation task into N local sub-tasks. In this new framework, we consider each neuron in the last hidden layer of the teacher network as a specialized sub-teacher. We also consider each neuron in the last hidden layer of the student network as a focused sub-student. We make each focused sub-student learn from one corresponding specialized sub-teacher and ignore the others. This will facilitate the task for the sub-student and keep it focused. Our proposed method is novel and can be combined with other distillation techniques. Empirical results show that our proposed approach outperforms the state-of-the-art methods by maintaining higher performance on most benchmark datasets. Furthermore, we propose a randomized variant of our approach, called Masked One-to-One Mapping. Rather than learning all the N sub-tasks simultaneously, we focus on learning a subset of these sub-tasks at each optimization step. This variant enables the student to digest the received flow of knowledge more effectively and yields superior results.
Cited by:	5
Resources:	[PDF] from aclanthology.org

:view less

Performance analysis of large language models in the domain of legal argument mining

2023

Abdullah Al Zubaer, Michael Granitzer, Jelena Mitrović
Frontiers in artificial intelligence 6, 1278796
:read more

Authors:	Abdullah Al Zubaer, Michael Granitzer, Jelena Mitrović
Publication date:	11/17/2023
Journal:	Frontiers in artificial intelligence
Volume:	6
Pages:	1278796
Publisher:	Frontiers Media SA
Description:	Generative pre-trained transformers (GPT) have recently demonstrated excellent performance in various natural language tasks. The development of ChatGPT and the recently released GPT-4 model has shown competence in solving complex and higher-order reasoning tasks without further training or fine-tuning. However, the applicability and strength of these models in classifying legal texts in the context of argument mining are yet to be realized and have not been tested thoroughly. In this study, we investigate the effectiveness of GPT-like models, specifically GPT-3.5 and GPT-4, for argument mining via prompting. We closely study the model's performance considering diverse prompt formulation and example selection in the prompt via semantic search using state-of-the-art embedding models from OpenAI and sentence transformers. We primarily concentrate on the argument component classification task on the legal corpus from the European Court of Human Rights. To address these models' inherent non-deterministic nature and make our result statistically sound, we conducted 5-fold cross-validation on the test set. Our experiments demonstrate, quite surprisingly, that relatively small domain-specific models outperform GPT 3.5 and GPT-4 in the F1-score for premise and conclusion classes, with 1.9% and 12% improvements, respectively. We hypothesize that the performance drop indirectly reflects the complexity of the structure in the dataset, which we verify through prompt and data analysis. Nevertheless, our results demonstrate a noteworthy variation in the performance of GPT models based on prompt formulation. We observe …
Cited by:	69
Resources:	[HTML] from frontiersin.org

:view less

A comprehensive dataset for webpage classification

2023

Mohammed Al-Maamari, Mahmoud Istaiti, Saber Zerhoudi, Michael Dinzinger, Michael Granit…
Open Search Symposium
:read more

Authors:	Mohammed Al-Maamari, Mahmoud Istaiti, Saber Zerhoudi, Michael Dinzinger, Michael Granitzer, Jelena Mitrović
Publication date:	10/04/2023
Journal:	Open Search Symposium
Description:	While webpage classification may not be a fundamental requirement for basic web crawling, it proves useful in enhancing the prioritization of crawled webpages. In this regard, our study presents a dataset of 116,000 URLs, complete with their content, specifically curated for webpage classification tasks. The primary goal of this research is to establish this comprehensive dataset with two levels of labels for URLs. Firstly, a broad level categorization dividing URLs into Malicious, Benign, or Adult, and secondly, a more nuanced labeling which includes 20 subclasses, providing a more granular view of the webpage content. The secondary objective is to leverage this dataset for testing and comparing the performance of various machine learning models, specifically Stochastic Gradient Descent (SGD) and Support Vector Classifier (SVC), in the task of webpage classification. This involves investigating the effectiveness of different input types (URLs only, raw HTML content, and parsed HTML content) and various tokenization methods (character-level, word-level, Byte Pair Encoding (BPE)[1]) on model performance. A total of 36 experiments were conducted, yielding several important findings. Using only the URL as input consistently resulted in the highest F1 score 0.94. Character-level tokenization consistently outperformed other tokenization techniques. There was a negligible difference in the accuracy of webpage classification between SGD and SVC models. This research’s findings demonstrate the viability of URL-based classification systems in web crawlers and shed light on optimal techniques for feature representation. The comprehensive …
Cited by:	2
Resources:	[PDF] from github.io

:view less

OWler: Preliminary results for building a Collaborative Open Web Crawler

2023

Michael Dinzinger, Michael and Al-Maamari, Mohammed and Zerhoudi, Saber and Istaiti, Ma…
Open Search Symposium 2023, 4-6 October 2023, CERN, Geneva, Switzerland (10.5281/zenodo…
:read more

Authors:	Michael Dinzinger, Michael and Al-Maamari, Mohammed and Zerhoudi, Saber and Istaiti, Mahmoud and Mitrović, Jelena and Granitzer
Publication date:	10/01/2023
Journal:	Open Search Symposium 2023, 4-6 October 2023, CERN, Geneva, Switzerland
Issue:	10.5281/zenodo.10581841
Cited by:	3

:view less

Deep Active Learning with Concept Drifts for Detection of Mercury’s Bow Shock and Magnetopause Crossings

2023

Sahib Julka, Rodion Ishmukhametov, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 390-404
:read more

Authors:	Sahib Julka, Rodion Ishmukhametov, Michael Granitzer
Publication date:	09/22/2023
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	390-404
Publisher:	Springer Nature Switzerland
Description:	Active learning has shown great potential for improving the efficiency of data annotation in scientific applications. In the field of planetary science, where large volumes of data are generated by remote sensing instruments, active learning can significantly reduce the cost and time required for data analysis. However, existing active learning (AL) methods for planetary science applications do not consider the potential for drifts in the data distribution, which can lead to model degradation over time. To address this issue, we propose a drift detection-based active learning approach for planetary science applications. The proposed approach uses a semi-supervised generative adversarial network (GAN) to detect concept drifts and an entropy-based AL sampling procedure to select the most informative orbits from each for training. We test this approach on the use case of detecting bow shock and magnetopause …
Cited by:	1

:view less

Gran is superior to graphrnn: node orderings, kernel-and graph embeddings-based metrics for graph generators

2023

Ousmane Touat, Julian Stier, Pierre-Edouard Portier, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 430-444
:read more

Authors:	Ousmane Touat, Julian Stier, Pierre-Edouard Portier, Michael Granitzer
Publication date:	09/22/2023
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	430-444
Publisher:	Springer Nature Switzerland
Description:	A wide variety of generative models for graphs have been proposed. They are used in drug discovery, road networks, neural architecture search, and program synthesis. Generating graphs has theoretical challenges, such as isomorphic representations – evaluating how well a generative model performs is difficult. Which model to choose depending on the application domain? We extensively study kernel-based metrics on distributions of graph invariants and manifold-based and kernel-based metrics in graph embedding space. Manifold-based metrics outperform kernel-based metrics in embedding space. We use these metrics to compare GraphRNN and GRAN, two well-known generative models for graphs, and unveil the influence of node orderings. It shows the superiority of GRAN over GraphRNN - further, our proposed adaptation of GraphRNN with a depth-first search ordering is effective for small-sized …
Cited by:	2
Resources:	[PDF] from arxiv.org

:view less

Knowledge distillation with segment anything (SAM) model for planetary geological mapping

2023

Sahib Julka, Michael Granitzer
International conference on machine learning, optimization, and data science, 68-77
:read more

Authors:	Sahib Julka, Michael Granitzer
Publication date:	09/22/2023
Journal:	International conference on machine learning, optimization, and data science
Pages:	68-77
Publisher:	Springer Nature Switzerland
Description:	Planetary science research involves analysing vast amounts of remote sensing data, which are often costly and time-consuming to annotate and process. One of the essential tasks in this field is geological mapping, which requires identifying and outlining regions of interest in planetary images, including geological features and landforms. However, manually labelling these images is a complex and challenging task that requires significant domain expertise and effort. To expedite this endeavour, we propose the use of knowledge distillation using the recently introduced cutting-edge Segment Anything (SAM) model. We demonstrate the effectiveness of this prompt-based foundation model for rapid annotation and quick adaptability to a prime use case of mapping planetary skylights. Our work reveals that with a small set of annotations obtained with the right prompts from the model and subsequently training a …
Cited by:	51
Resources:	[PDF] from arxiv.org

:view less

Machine learning system for various computer applications

2023

:read more

Publication date:	09/19/2023
Description:	The object of the present invention is thus to propose a system for detecting fraud during an identification, making it possible to overcome at least some of the disadvantages of the prior art, by proposing a machine learning system for various computer applications enabling text mining to detect defects or anomalies in a authentication, operation or transaction carried out by the application comprising: A hardware and software arrangement forming a preprocessing system;
Cited by:	6
Resources:	[PDF] from googleapis.com

:view less

A dataset of German legal reference annotations

2023

Harshil Darji, Jelena Mitrović, Michael Granitzer
Proceedings of the Nineteenth International Conference on Artificial Intelligence and L…
:read more

Authors:	Harshil Darji, Jelena Mitrović, Michael Granitzer
Publication date:	06/19/2023
Journal:	Proceedings of the Nineteenth International Conference on Artificial Intelligence and Law
Pages:	392-396
Description:	The field of legal Natural Language Processing faces a lot of challenges due to the unavailability of properly structured datasets. One such instance is the need for a dataset that not only separates different parts of legal references, such as an article or paragraph number but also provides information about what a particular legal reference dictates. Having access to such a dataset can provide easy access to researchers working on experiments such as context similarity between law texts and legal cases that refer to a particular law. In this paper, we present a dataset of 2944 legal references in German law that are manually annotated by law experts. This dataset has 21 properties for each law reference in the dataset, such as Buch, Teil, Titel, Untertitel, etc. It also provides the complete text of each law reference in the dataset, along with specific paragraph text mentioned in the law reference. Furthermore, using this …
Cited by:	11
Resources:	[PDF] from acm.org

:view less

Technical Report: Impact of Position Bias on Language Models in Token Classification

2023

Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović
arXiv preprint arXiv:2304.13567
:read more

Authors:	Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović
Publication date:	04/26/2023
Journal:	arXiv preprint arXiv:2304.13567
Description:	Language Models (LMs) have shown state-of-the-art performance in Natural Language Processing (NLP) tasks. Downstream tasks such as Named Entity Recognition (NER) or Part-of-Speech (POS) tagging are known to suffer from data imbalance issues, particularly regarding the ratio of positive to negative examples and class disparities. This paper investigates an often-overlooked issue of encoder models, specifically the position bias of positive examples in token classification tasks. For completeness, we also include decoders in the evaluation. We evaluate the impact of position bias using different position embedding techniques, focusing on BERT with Absolute Position Embedding (APE), Relative Position Embedding (RPE), and Rotary Position Embedding (RoPE). Therefore, we conduct an in-depth evaluation of the impact of position bias on the performance of LMs when fine-tuned on token classification benchmarks. Our study includes CoNLL03 and OntoNote5.0 for NER, English Tree Bank UD_en, and TweeBank for POS tagging. We propose an evaluation approach to investigate position bias in transformer models. We show that LMs can suffer from this bias with an average drop ranging from 3% to 9% in their performance. To mitigate this effect, we propose two methods: Random Position Shifting and Context Perturbation, that we apply on batches during the training process. The results show an improvement of 2% in the performance of the model on CoNLL03, UD_en, and TweeBank.
Resources:	[PDF] from arxiv.org

:view less

German BERT model for legal named entity recognition

2023

Harshil Darji, Jelena Mitrović, Michael Granitzer
arXiv preprint arXiv:2303.05388
:read more

Authors:	Harshil Darji, Jelena Mitrović, Michael Granitzer
Publication date:	03/07/2023
Journal:	arXiv preprint arXiv:2303.05388
Description:	The use of BERT, one of the most popular language models, has led to improvements in many Natural Language Processing (NLP) tasks. One such task is Named Entity Recognition (NER) i.e. automatic identification of named entities such as location, person, organization, etc. from a given text. It is also an important base step for many NLP tasks such as information extraction and argumentation mining. Even though there is much research done on NER using BERT and other popular language models, the same is not explored in detail when it comes to Legal NLP or Legal Tech. Legal NLP applies various NLP techniques such as sentence similarity or NER specifically on legal data. There are only a handful of models for NER tasks using BERT language models, however, none of these are aimed at legal documents in German. In this paper, we fine-tune a popular BERT language model trained on German data (German BERT) on a Legal Entity Recognition (LER) dataset. To make sure our model is not overfitting, we performed a stratified 10-fold cross-validation. The results we achieve by fine-tuning German BERT on the LER dataset outperform the BiLSTM-CRF+ model used by the authors of the same LER dataset. Finally, we make the model openly available via HuggingFace.
Cited by:	77
Resources:	[PDF] from arxiv.org

:view less

FaDE: Fast DARTS Estimator on Hierarchical NAS Spaces

2023

Simon Neumeyer, Julian Stier, Michael Granitzer
OPT 2023: Optimization for Machine Learning
:read more

Authors:	Simon Neumeyer, Julian Stier, Michael Granitzer
Publication date:	01/01/2023
Journal:	OPT 2023: Optimization for Machine Learning
Description:	Vast search spaces and expensive architecture evaluations make neural architecture search a challenging problem. Hierarchical search spaces allow for comparatively cheap evaluations on neural network sub modules to serve as surrogate for architecture evaluations. Yet, sometimes the hierarchy is too restrictive or the surrogate fails to generalize. We present FaDE that uses differentiable architecture search to iteratively obtain relative performance predictions on finite regions of a hierarchical neural architecture search space. The relative nature of these ranks calls for a memory-less, batch-wise outer search algorithm which we provide in form of an evolutionary approach that features a pseudo-gradient descent. FaDE is especially suited on deep hierarchical respectively multi-cell search spaces which it can explore by linear instead of exponential cost and therefore eliminates the need for a proxy search space. FaDE solely trains on the neural architecture search space, not on any space of neural architecture sub modules. Our experiments show that firstly, FaDE-ranks on finite regions of the search space correlate with corresponding architecture performances and secondly, the ranks can empower a pseudo-gradient evolutionary search on the complete neural architecture search space.
Cited by:	1
Resources:	[PDF] from openreview.net

:view less

ESTHER: Ontology of Rhetorical Figures in English.

2023

Ramona Kühn, Jelena Mitrovic, Michael Granitzer
JOWO
:read more

Authors:	Ramona Kühn, Jelena Mitrovic, Michael Granitzer
Publication date:	01/01/2023
Journal:	JOWO
Description:	ESTHER, the English ontology of rhetorical figures, is a formal domain ontology of rhetorical figures in the English language. Rhetorical figures convey a subtle meaning or sentiment that often cannot be grasped by modern sentiment analyzers, machine translation systems, or text summarizers due to the lack of annotated datasets and machine-readable formats. Our ontology is a formal documentation of rhetorical figures to improve various natural language processing tasks. Existing ontologies in this domain mainly focus on figures in other languages, such as Serbian or German. Our ESTHER ontology fills this gap by modeling 85 English rhetorical figures. It aligns with existing ontologies to allow interoperability with other languages. In addition, the ESTHER ontology includes hierarchical relations and compositional features of figures to model their dependency relations. Besides identifying the most important steps of the good practice of ontology modeling, we also provide an overview of current tools and methods of ontological evaluation, validation, and documentation for the ESTHER ontology. The ontology does not only serve as a knowledge base and documentation of rhetorical figures but also has a practical use case. We developed a tool to show how the ontology helps annotators without linguistic understanding to annotate instances of rhetorical figures.
Cited by:	4
Resources:	[PDF] from ceur-ws.org

:view less

Hidden in plain sight: Can german wiktionary and wordnets facilitate the detection of antithesis?

2023

Ramona Kühn, Jelena Mitrović, Michael Granitzer
Proceedings of the 12th global wordnet conference, 106-116
:read more

Authors:	Ramona Kühn, Jelena Mitrović, Michael Granitzer
Publication date:	01/01/2023
Journal:	Proceedings of the 12th global wordnet conference
Pages:	106-116
Description:	Existing wordnets mainly focus on synonyms, while antonyms have often been neglected, especially in wordnets in languages other than English. In this paper, we show how regular expressions are used to generate an antonym resource for German by using Wiktionary as a source. This resource contains antonyms for 45499 words. The antonyms can be used to extend existing wordnets. We show that this is important by comparing our antonym resource to the antonyms in OdeNet, the only freely available German wordnet that contains antonyms for 3059 words. We demonstrate that antonyms are relevant for the detection of the rhetorical figure antithesis. This figure has been known to influence the audience by creating contradiction and using a parallel sentence structure combined with antonyms. We first detect parallelism with part-of-speech tags and then apply our rule-based antithesis detection algorithm to a dataset of the messenger service Telegram. We evaluate our approach and achieve a precision of 57% and a recall of 45% thus overcoming the existing approaches.
Cited by:	9
Resources:	[PDF] from aclanthology.org

:view less

Lessons learned from the 1st Ariel Machine Learning Challenge: Correcting transiting exoplanet light curves for stellar spots

2023

Nikolaos Nikolaou, Ingo P Waldmann, Angelos Tsiaras, Mario Morvan, Billy Edwards, Kai H…
RAS Techniques and Instruments 2 (1), 695-709
:read more

Authors:	Nikolaos Nikolaou, Ingo P Waldmann, Angelos Tsiaras, Mario Morvan, Billy Edwards, Kai Hou Yip, Alexandra Thompson, Giovanna Tinetti, Subhajit Sarkar, James M Dawson, Vadim Borisov, Gjergji Kasneci, Matej Petković, Tomaž Stepišnik, Tarek Al-Ubaidi, Rachel Louise Bailey, Michael Granitzer, Sahib Julka, Roman Kern, Patrick Ofner, Stefan Wagner, Lukas Heppe, Mirko Bunse, Katharina Morik, Luís F Simões
Publication date:	01/01/2023
Journal:	RAS Techniques and Instruments
Volume:	2
Issue:	1
Pages:	695-709
Publisher:	Oxford University Press
Description:	The last decade has witnessed a rapid growth of the field of exoplanet discovery and characterization. However, several big challenges remain, many of which could be addressed using machine learning methodology. For instance, the most prolific method for detecting exoplanets and inferring several of their characteristics, transit photometry, is very sensitive to the presence of stellar spots. The current practice in the literature is identifying the effects of spots visually and correcting them manually or discarding the affected data. This paper explores a first step towards fully automating the efficient and precise derivation of transit depths from transit light curves in the presence of stellar spots. The primary focus of the paper is to present in detail a diverse arsenal of methods for doing so. The methods and results we present were obtained in the context of the 1st Machine Learning Challenge organized for the …
Cited by:	19
Resources:	[HTML] from oup.com Full View

:view less

“The Need for Speed”: Extracting Session Keys From the Main Memory Using Brute-force and Machine Learning

2022

Stewart Sentanoe, Christofer Fellicious, Hans P Reiser, Michael Granitzer
2022 IEEE International Conference on Trust, Security and Privacy in Computing and Comm…
:read more

Authors:	Stewart Sentanoe, Christofer Fellicious, Hans P Reiser, Michael Granitzer
Publication date:	12/09/2022
Journal:	2022 IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)
Pages:	1028-1035
Publisher:	IEEE
Description:	Digital forensics has become an important topic during this digital devices era. One major problem is extracting critical information from a digital device, and one of such data sources from a digital device is the main memory. The main memory holds many data that can be as simple as a text and a number or as complex as a data structure that holds cryptography keys. In line with the growth of the digital devices era, the need for privacy is also becoming an emerging topic. Encryption is one of the ways to achieve privacy during data transmission. As mentioned, the main memory might hold the session keys to encrypt or decrypt such secure data transmissions. This paper proposes a pure brute-force and a machine learning augmented brute force method to extract session keys from the main memory. In addition, we reduce the training data footprint using a novel entropy-based preprocessing method. We choose the …
Cited by:	8

:view less

Method for intrusion detection to detect malicious insider threat activities and system for intrusion detection

2022

:read more

Publication date:	12/08/2022
Description:	A method and system for intrusion detection to detect malicious insider threat activities within a network user profiles. The method includes determining a behavior pat tern for each user profile based on activity events, wherein the determination of the behavior pattern is executed by a Recurrent Neural Network. The method includes determin ing normal activity events and abnormal activity events for each user profile based on the behavior patterns, wherein the determination of the normal activity events and the abnor mal activity events is executed by a Feed-Forward Neural Network. The method includes evaluating whether a recorded activity event is a normal activity event or an abnormal activity event based on the behavior pattern and the determined normal activity events and abnormal events for that user profile. The method includes detecting mali cious activity for the user profile, if the recorded activity event is …
Cited by:	7
Resources:	[PDF] from googleapis.com

:view less

The simiir 2.0 framework: User types, markov model-based interaction simulation, and advanced query generation

2022

Saber Zerhoudi, Sebastian Günther, Kim Plassmeier, Timo Borst, Christin Seifert, Matthi…
Proceedings of the 31st ACM International Conference on Information & Knowledge Managem…
:read more

Authors:	Saber Zerhoudi, Sebastian Günther, Kim Plassmeier, Timo Borst, Christin Seifert, Matthias Hagen, Michael Granitzer
Publication date:	10/17/2022
Journal:	Proceedings of the 31st ACM International Conference on Information & Knowledge Management
Pages:	4661-4666
Description:	Simulated user retrieval system interactions enable studies with controlled user behavior. To this end, the SimIIR framework offers static, rule-based methods. We present an extended SimIIR 2.0 version with new components for dynamic user type-specific Markov model-based interactions and more realistic query generation. A flexible modularization ensures that the SimIIR 2.0 framework can serve as a platform to implement, combine, and run the growing number of proposed search behavior and query simulation ideas.
Cited by:	36
Resources:	[PDF] from acm.org

:view less

Expressive and Intuitive Models for Automated Context Representation Learning in Credit-Card Fraud Detection

2022

Kanishka Ghosh Dastidar, Wissam Siblini, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 154-168
:read more

Authors:	Kanishka Ghosh Dastidar, Wissam Siblini, Michael Granitzer
Publication date:	09/19/2022
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	154-168
Publisher:	Springer Nature Switzerland
Description:	The fraud detection literature unanimously shows that the use of a cardholder’s transaction history as context improves the classification of the current transaction. Context representation is usually performed through either of two approaches. The first, manual feature engineering, is expensive, restricted, and hard to maintain as it relies on human expertise. The second, automatic context representation, removes the human dependency by learning new features directly on the fraud data with an end-to-end neural network. The LSTM and the more recent Neural Feature Aggregate Generator (NAG) are examples of such an approach. The architecture of the NAG is inspired by manual feature aggregates and addresses several of their limitations, primarily because it is automatic. However, it still has several drawbacks that we aim to address in this paper. In particular, we propose to extend the NAG in the following two …
Cited by:	1

:view less

Pooling Graph Convolutional Networks for Structural Performance Prediction

2022

Lorenz Wendlinger, Michael Granitzer, Christofer Fellicious
International Conference on Machine Learning, Optimization, and Data Science, 1-16
:read more

Authors:	Lorenz Wendlinger, Michael Granitzer, Christofer Fellicious
Publication date:	09/19/2022
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	1-16
Publisher:	Springer Nature Switzerland
Description:	Neural Architecture Search can help in finding high-performance task specific neural network architectures. However, the training of architectures that is required for fitness computation can be prohibitively expensive. Employing surrogate models as performance predictors can reduce or remove the need for these costly evaluations. We present a deep graph learning approach that achieves state-of-the-art performance in multiple NAS performance prediction benchmarks. In contrast to other methods, this model is purely supervised, which can be a methodologic advantage, as it does not rely on unlabeled instances sampled from complex search spaces.
Cited by:	1

:view less

Neural network based drift detection

2022

Christofer Fellicious, Lorenz Wendlinger, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 370-383
:read more

Authors:	Christofer Fellicious, Lorenz Wendlinger, Michael Granitzer
Publication date:	09/19/2022
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	370-383
Publisher:	Springer Nature Switzerland
Description:	The unprecedented growth in machine learning has shed light on its unique set of challenges. One such challenge is apparent changes in the input data distribution over time known as Concept Drifts. In such cases, the model’s performance degrades according to the changes in the data distribution. The remedy for concept drifts is retraining the model with the most recent data to improve the model’s performance. The significant issue is identifying the precise point at which the model must be updated for maximum performance benefits with minimum retraining effort. This problem is challenging to address in unsupervised detection methods with no access to label data to identify the changing distributions for the targets of the input data. Here, we present our unsupervised method based on a Generative Adversarial Network and a feed forward neural network for detecting concept drifts without the need for target …
Cited by:	7

:view less

Deep active learning for detection of mercury’s bow shock and magnetopause crossings

2022

Sahib Julka, Nikolas Kirschstein, Michael Granitzer, Alexander Lavrukhin, Ute Amerstorfer
Joint European conference on machine learning and knowledge discovery in databases, 452…
:read more

Authors:	Sahib Julka, Nikolas Kirschstein, Michael Granitzer, Alexander Lavrukhin, Ute Amerstorfer
Publication date:	09/19/2022
Journal:	Joint European conference on machine learning and knowledge discovery in databases
Pages:	452-467
Publisher:	Springer Nature Switzerland
Description:	Accurate and timely detection of bow shock and magnetopause crossings is essential for understanding the dynamics of a planet’s magnetosphere. However, for Mercury, due to the variable nature of its magnetosphere, this remains a challenging task. Existing approaches based on geometric equations only provide average boundary shapes, and can be hard to generalise to environments with variable conditions. On the other hand, data-driven methods require large amounts of annotated data to account for variations, which can scale up the costs quickly. We propose to solve this problem with machine learning. To this end, we introduce a suitable dataset, prepared by processing raw measurements from NASA’s MESSENGER (MErcury Surface, Space Environment, GEochemistry, and Ranging) mission and design a five-class supervised learning problem. We perform an architectural search to find a suitable …
Cited by:	8
Resources:	[PDF] from researchgate.net

:view less

Simulating user querying behavior using embedding space alignment

2022

Saber Zerhoudi, Michael Granitzer
International Conference on Theory and Practice of Digital Libraries, 386-394
:read more

Authors:	Saber Zerhoudi, Michael Granitzer
Publication date:	09/15/2022
Journal:	International Conference on Theory and Practice of Digital Libraries
Pages:	386-394
Publisher:	Springer International Publishing
Description:	Simulation is used as a cost-efficient and repeatable means of experimentation to support Information Retrieval (IR) systems and digital libraries with more realistic directives when user interaction data is lacking. While simulation has been mainly used for the study of user search behaviours, we argue that the potential of using it for generating querying behavior is in its infancy. In this paper, we describe a pilot study to simulate query sessions via embedding alignment techniques and assess its capability of transferring knowledge from one source to the other. Besides, we validate the simulated query sessions using a cost and benefit evaluation task and investigate whether they can reproduce properties of real query sessions. We found that Skip-thought vectors are well suited for query embedding. In particular, using Skip-thought with vocabulary expansion as a support for embedding alignment techniques is …
Cited by:	3

:view less

SmartKex: machine learning assisted SSH keys extraction from the heap dump

2022

Christofer Fellicious, Stewart Sentanoe, Michael Granitzer, Hans P Reiser
arXiv preprint arXiv:2209.05243
:read more

Authors:	Christofer Fellicious, Stewart Sentanoe, Michael Granitzer, Hans P Reiser
Publication date:	09/12/2022
Journal:	arXiv preprint arXiv:2209.05243
Description:	Digital forensics is the process of extracting, preserving, and documenting evidence in digital devices. A commonly used method in digital forensics is to extract data from the main memory of a digital device. However, the main challenge is identifying the important data to be extracted. Several pieces of crucial information reside in the main memory, like usernames, passwords, and cryptographic keys such as SSH session keys. In this paper, we propose SmartKex, a machine-learning assisted method to extract session keys from heap memory snapshots of an OpenSSH process. In addition, we release an openly available dataset and the corresponding toolchain for creating additional data. Finally, we compare SmartKex with naive brute-force methods and empirically show that SmartKex can extract the session keys with high accuracy and high throughput. With the provided resources, we intend to strengthen the research on the intersection between digital forensics, cybersecurity, and machine learning.
Cited by:	10
Resources:	[PDF] from arxiv.org

:view less

Reconciliation of Mental Concepts with Graph Neural Networks

2022

Lorenz Wendlinger, Gerd Hübscher, Andreas Ekelhart, Michael Granitzer
International Conference on Database and Expert Systems Applications, 133-146
:read more

Authors:	Lorenz Wendlinger, Gerd Hübscher, Andreas Ekelhart, Michael Granitzer
Publication date:	07/29/2022
Journal:	International Conference on Database and Expert Systems Applications
Pages:	133-146
Publisher:	Springer International Publishing
Description:	In the digital age, knowledge processes can be formalized and simplified using task management systems. As they evolve, so must the underlying schemata to retain harmony and concurrency with the real world. In this work we present a graph neural network model that can help in reconciling these data. It can do so by leveraging a novel propagation rule that does not presume reciprocal dependency but is able to represent it still. Thereby it can predict structures in the form of usage links with high accuracy and assist in the reconstruction of missing information. We evaluate this model on a new knowledge management dataset and show that it is superior to traditional embedding methods. Further, we show that it outperforms related work in an established general link prediction task.

:view less

Automatic detection of bow shock and magnetopause boundaries at Mercury using MESSENGER magnetometer data

2022

Dmitrii Nevskii, Alexander Lavrukhin, Sahib Julka, David Parunakian, Michael Granitzer
44th COSPAR Scientific Assembly. Held 16-24 July 44, 475
:read more

Authors:	Dmitrii Nevskii, Alexander Lavrukhin, Sahib Julka, David Parunakian, Michael Granitzer
Publication date:	07/01/2022
Journal:	44th COSPAR Scientific Assembly. Held 16-24 July
Volume:	44
Pages:	475
Description:	The aim of this work is to develop methods for automatic processing of the MESSENGER (MErcury Surface, Space ENvironment, GEochemistry, and Ranging) spacecraft magnetometer data to search for the position of the bow shock and the magnetopause of the Mercury's magnetosphere. In our work we use different methods, including neural networks. In 2011-2015 MESSENGER spacecraft completed more than 4000 orbits around Mercury, thus giving data of more than 8000 crossings of bow shock as well as magnetopause of the planet. Given large data, Neural Networks can be expected to approximate complex functions, which often surpass deterministic and rule-based methods, in a variety of time series tasks like classification, time series forecasting and rare time series event detection. We leverage these to develop a predictor, that can be used in real-time during orbit, to predict magnetic region for each …

:view less

Network analysis of german covid-19 related discussions on telegram

2022

Valentin Peter, Ramona Kühn, Jelena Mitrović, Michael Granitzer, Hannah Schmid-Petri
International Conference on Applications of Natural Language to Information Systems, 25-32
:read more

Authors:	Valentin Peter, Ramona Kühn, Jelena Mitrović, Michael Granitzer, Hannah Schmid-Petri
Publication date:	06/13/2022
Journal:	International Conference on Applications of Natural Language to Information Systems
Pages:	25-32
Publisher:	Springer International Publishing
Description:	We present an effective way to create a dataset from relevant channels and groups of the messenger service Telegram, to detect clusters in this network, and to find influential actors. Our focus lies on the network of German COVID-19 sceptics that formed on Telegram along with growing restrictions meant to prevent the spreading of COVID-19. We create the dataset by using a scraper based on exponential discriminative snowball sampling, combining two different approaches. We show the best way to define a starting point for the sampling and to detect relevant neighbouring channels for the given data. Community clusters in the network are detected by using the Louvain method. Furthermore, we show influential channels and actors by defining a PageRank based ranking scheme. A heatmap illustrates the correlation between the number of channel members and the ranking. We also examine the growth of the …
Cited by:	14
Resources:	[PDF] from google.com

:view less

GRhOOT: Ontology of rhetorical figures in German

2022

Ramona Kühn, Jelena Mitrović, Michael Granitzer
Proceedings of the thirteenth language resources and evaluation conference, 4001-4010
:read more

Authors:	Ramona Kühn, Jelena Mitrović, Michael Granitzer
Publication date:	06/01/2022
Journal:	Proceedings of the thirteenth language resources and evaluation conference
Pages:	4001-4010
Description:	GRhOOT, the German RhetOrical OnTology, is a domain ontology of 110 rhetorical figures in the German language. The overall goal of building an ontology of rhetorical figures in German is not only the formal representation of different rhetorical figures, but also allowing for their easier detection, thus improving sentiment analysis, argument mining, detection of hate speech and fake news, machine translation, and many other tasks in which recognition of non-literal language plays an important role. The challenge of building such ontologies lies in classifying the figures and assigning adequate characteristics to group them, while considering their distinctive features. The ontology of rhetorical figures in the Serbian language was used as a basis for our work. Besides transferring and extending the concepts of the Serbian ontology, we ensured completeness and consistency by using description logic and SPARQL queries. Furthermore, we show a decision tree to identify figures and suggest a usage scenario on how the ontology can be utilized to collect and annotate data.
Cited by:	16
Resources:	[PDF] from aclanthology.org

:view less

The importance of future information in credit card fraud detection

2022

Kanishka Ghosh Dastidar, Michael Granitzer, Wissam Siblini
International Conference on Artificial Intelligence and Statistics, 10067-10077
:read more

Authors:	Kanishka Ghosh Dastidar, Michael Granitzer, Wissam Siblini
Publication date:	05/03/2022
Journal:	International Conference on Artificial Intelligence and Statistics
Pages:	10067-10077
Publisher:	PMLR
Description:	Fraud detection systems (FDS) mainly perform two tasks:(i) real-time detection while the payment is being processed and (ii) posterior detection to block the card retrospectively and avoid further frauds. Since human verification is often necessary and the payment processing time is limited, the second task manages the largest volume of transactions. In the literature, fraud detection challenges and algorithms performance are widely studied but the very formulation of the problem is never disrupted: it aims at predicting if a transaction is fraudulent based on its characteristics and the past transactions of the cardholder. Yet, in posterior detection, verification often takes days, so new payments on the card become available before a decision is taken. This is our motivation to propose a new paradigm: posterior fraud detection with" future" information. We start by providing evidence of the on-time availability of subsequent transactions, usable as extra context to improve detection. We then design a Bidirectional LSTM to make use of these transactions. On a real-world dataset with over 30 million transactions, it achieves higher performance than a regular LSTM, which is the state-of-the-art classifier for fraud detection that only uses the past context. We also introduce new metrics to show that the proposal catches more frauds, more compromised cards, and based on their earliest frauds. We believe that future works on this new paradigm will have a significant impact on the detection of compromised cards.
Cited by:	26
Resources:	[PDF] from mlr.press

:view less

Evaluating simulated user interaction and search behaviour

2022

Saber Zerhoudi, Michael Granitzer, Christin Seifert, Joerg Schloetterer
European Conference on Information Retrieval, 240-247
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Christin Seifert, Joerg Schloetterer
Publication date:	04/05/2022
Journal:	European Conference on Information Retrieval
Pages:	240-247
Publisher:	Springer International Publishing
Description:	Simulating user sessions in a way that comes closer to the original user interactions is key to generating user data at any desired volume and variety such that A/B-testing in domain-specific search engines becomes scalable. In recent years, research on evaluating Information Retrieval (IR) systems has mainly focused on simulation as means to improve users models and evaluation metrics about the performance of search engines using test collections and user studies. However, test collections contain no user interaction data and user studies are expensive to conduct. Thus there is a need in developing a methodology for evaluating simulated user sessions. In this paper, we propose evaluation metrics to assess the realism of simulated sessions and describe a pilot study to assess the capability of generating simulated search sequences representing an approximation of real behaviour. Our findings highlight the …
Cited by:	9
Resources:	[PDF] from ecir2022.org

:view less

The Importance of Future Information in Credit Card Fraud Detection

2022

Van Bach Nguyen, Kanishka Ghosh Dastidar, Michael Granitzer, Wissam Siblini
arXiv e-prints, arXiv: 2204.05265
:read more

Authors:	Van Bach Nguyen, Kanishka Ghosh Dastidar, Michael Granitzer, Wissam Siblini
Publication date:	04/01/2022
Journal:	arXiv e-prints
Pages:	arXiv: 2204.05265
Description:	Fraud detection systems (FDS) mainly perform two tasks:(i) real-time detection while the payment is being processed and (ii) posterior detection to block the card retrospectively and avoid further frauds. Since human verification is often necessary and the payment processing time is limited, the second task manages the largest volume of transactions. In the literature, fraud detection challenges and algorithms performance are widely studied but the very formulation of the problem is never disrupted: it aims at predicting if a transaction is fraudulent based on its characteristics and the past transactions of the cardholder. Yet, in posterior detection, verification often takes days, so new payments on the card become available before a decision is taken. This is our motivation to propose a new paradigm: posterior fraud detection with" future" information. We start by providing evidence of the on-time availability of subsequent …

:view less

Simulating User Interaction and Search Behaviour in Digital Libraries.

2022

Saber Zerhoudi, Michael Granitzer, Christin Seifert, Jörg Schlötterer
IRCDL
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Christin Seifert, Jörg Schlötterer
Publication date:	02/24/2022
Journal:	IRCDL
Description:	Providing a satisfying user experience is key to successful digital libraries. An efficient search user interface design requires a detailed understanding of user behaviour to match their needs. A common approach to modelling users is to transform the recorded users’ navigation behaviour to Markov models and analyse them to provide personalised content. However, personalisation based on log data is hard to achieve due to the lack of sufficient daily user interactions in comparison to web search engines, even when achieved, it’s usually on the cost of users’ privacy and the confidentiality of their personal information potentially leading to privacy violation. In order to allow data analysis while retaining users’ privacy, we propose a Markov approach to simulate user sessions and help reducing the amount of collected user data while preserving the profiling efficiency. Specifically, given log data of search sequences performed by users in a digital library, we explore basic, conditional, contextual, time-aware and query-change Markov models to simulate similar search sequences. Our results on an academic search engine log dataset show that our methods reliably simulate global and type specific search behaviour. Simulated search sequences representing an approximation of real behaviour can be used to generate log data at any desired volume and variety such that A/B-testing digital libraries becomes scalable.
Cited by:	15
Resources:	[PDF] from ceur-ws.org

:view less

Deepstruct–linking deep learning and graph theory

2022

Julian Stier, Michael Granitzer
Software Impacts 11, 100193
:read more

Authors:	Julian Stier, Michael Granitzer
Publication date:	02/01/2022
Journal:	Software Impacts
Volume:	11
Pages:	100193
Publisher:	Elsevier
Description:	deepstruct connects deep learning models and graph theory such that different graph structures can be imposed on neural networks or graph structures can be extracted from trained neural network models. For this, deepstruct provides deep neural network models with different restrictions which can be created based on an initial graph. Further, tools to extract graph structures from trained models are available. This step of extracting graphs can be computationally expensive even for models of just a few dozen thousand parameters and poses a challenging problem.deepstruct supports research in pruning, neural architecture search, automated network design and structure analysis of neural networks.
Cited by:	5
Resources:	[HTML] from sciencedirect.com

:view less

Towards a Unified Multilingual Ontology for Rhetorical Figures.

2022

Yetian Wang, Ramona Kühn, Randy Allen Harris, Jelena Mitrovic, Michael Granitzer
KDIR, 117-127
:read more

Authors:	Yetian Wang, Ramona Kühn, Randy Allen Harris, Jelena Mitrovic, Michael Granitzer
Publication date:	01/01/2022
Journal:	KDIR
Pages:	117-127
Description:	Formal ontologies for rhetorical figures have been developed to improve the computational detection for different applications in the area of Natural Language Processing, such as hate speech and fake news detection, argumentation mining, and sentiment analysis. The existing ontologies all model different aspects of rhetorical figures, thus creating a variety of formalisms and in the worst case, creating incompatibilities and contradictory representations. In this paper, we focus on figures of perfect lexical repetition and their representation in three ontologies in three different languages: The Ploke ontology, the Serbian RetFig, and the German GRhOOT ontology. We combine those ontologies to benefit from synergy effects and create a multilingual, coherent, robust, and modular ontology for rhetorical figures of perfect lexical repetition.
Cited by:	9
Resources:	[PDF] from scitepress.org

:view less

Deep Active Learning for Detection of Mercury’s Bow Shock and Magnetopause Crossings

2022

U.V. Julka, S., Kirschstein, N., Granitzer, M., Lavrukhin, A.S., & Amerstorfer
European Conference on Machine Learning and Principles and Practice of Knowledge Discov…
:read more

Authors:	U.V. Julka, S., Kirschstein, N., Granitzer, M., Lavrukhin, A.S., & Amerstorfer
Publication date:	01/01/2022
Journal:	European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD)

:view less

Conditional Generative Adversarial Networks for Speed Control in Trajectory Simulation

2022

Michael Granitzer Sahib Julka, Vishal Sowrirajan, Joerg Schloetterer
Machine Learning, Optimization, and Data Science,, 436--450
:read more

Authors:	Michael Granitzer Sahib Julka, Vishal Sowrirajan, Joerg Schloetterer
Publication date:	01/01/2022
Journal:	Machine Learning, Optimization, and Data Science,
Pages:	436--450
Publisher:	Springer International Publishing
Cited by:	12

:view less

Query change as a contextual Markov model for simulating user search behaviour

2021

Saber Zerhoudi, Michael Granitzer, Jörg Schlötterer, Christin Seifert
Proceedings of the 13th Annual Meeting of the Forum for Information Retrieval Evaluatio…
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Jörg Schlötterer, Christin Seifert
Publication date:	12/13/2021
Journal:	Proceedings of the 13th Annual Meeting of the Forum for Information Retrieval Evaluation
Pages:	43-51
Description:	Search engine users issue queries to formulate their information need and gain useful insights. However, it is challenging for search engines to understand different users’ search type intents and return appropriate results. Simulating user search behaviour allows information retrieval systems (IR) to parameterise the a-priori distribution of search types using different back-end configurations and user interface variants to improve the retrieval functionality. In this paper, we propose a formal Markov approach in which we utilise the context discovery process to model user-type specific behaviour by capturing the user’s query change in a search session. Contextual Markov models have been used in the past to improve the prediction of user intentions, we investigate here their efficiency in simulating user-type specific interactions. Additionally, we provide an empirical and classification-based evaluation that can be used …
Cited by:	15
Resources:	[PDF] from acm.org

:view less

Exploring semantic similarity between German legal texts and referred laws

2021

Harshil Darji, Jelena Mitrović, Michael Granitzer
International Joint Conference on Knowledge Discovery, Knowledge Engineering, and Knowl…
:read more

Authors:	Harshil Darji, Jelena Mitrović, Michael Granitzer
Publication date:	10/25/2021
Journal:	International Joint Conference on Knowledge Discovery, Knowledge Engineering, and Knowledge Management
Pages:	37-50
Publisher:	Springer International Publishing
Description:	The calculation of semantic similarity is an important task in Natural Language Processing (NLP). There is a growing interest in this task in the research community, especially following the advent of new, ever-evolving neural architectures. However, this technique has not been explored in-depth in the realm of automatic processing of legal data, the area we often call Legal NLP or Legal Tech. In this paper, we aim to use semantic similarity to identify the relations between legal sentences that refer to a certain law and the law text itself. The semantic similarity score is calculated using cosine similarity between sentence embeddings. In our work, we use sentence transformers to get the sentence embeddings for our legal text. The results we achieve by using two separate sentence transformers, Cross English & German RoBERTa and all-MiniLM-L6-v2, provide a semantic similarity score of approximately 0.45 and 0.4 …
Cited by:	10
Resources:	[PDF] from google.com

:view less

Experiments on properties of hidden structures of sparse neural networks

2021

Julian Stier, Harshil Darji, Michael Granitzer
International Conference on Machine Learning, Optimization, and Data Science, 380-394
:read more

Authors:	Julian Stier, Harshil Darji, Michael Granitzer
Publication date:	10/04/2021
Journal:	International Conference on Machine Learning, Optimization, and Data Science
Pages:	380-394
Publisher:	Springer International Publishing
Description:	Sparsity in the structure of Neural Networks can lead to less energy consumption, less memory usage, faster computation times on convenient hardware, and automated machine learning. If sparsity gives rise to certain kinds of structure, it can explain automatically obtained features during learning. We provide insights into experiments in which we show how sparsity can be achieved through prior initialization, pruning, and during learning, and answer questions on the relationship between the structure of Neural Networks and their performance. This includes the first work of inducing priors from network theory into Recurrent Neural Networks and an architectural performance prediction during a Neural Architecture Search. Within our experiments, we show how magnitude class blinded pruning achieves 97.5% on MNIST with 80% compression and re-training, which is 0.5 points more than without compression, that …
Cited by:	7
Resources:	[PDF] from arxiv.org

:view less

Methods for automatic machine-learning workflow analysis

2021

Lorenz Wendlinger, Emanuel Berndl, Michael Granitzer
Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 52-67
:read more

Authors:	Lorenz Wendlinger, Emanuel Berndl, Michael Granitzer
Publication date:	09/10/2021
Journal:	Joint European Conference on Machine Learning and Knowledge Discovery in Databases
Pages:	52-67
Publisher:	Springer International Publishing
Description:	Developing real-world Machine Learning-based Systems goes beyond algorithm development. ML algorithms are usually embedded in complex pre-processing steps and consider different stages like development, testing or deployment. Managing workflows poses several challenges, such as workflow versioning, sharing pipeline elements or optimizing individual workflow elements - tasks which are usually conducted manually by data scientists. A dataset containing 16 035 real-world Machine Learning and Data Science Workflows extracted from the ONE DATA platform (https://onelogic.de/en/one-data/) is explored and made available. Based on our analysis, we develop a representation learning algorithm using a graph-level Graph Convolutional Network with explicit residuals which exploits workflow versioning history. Moreover, this method can easily be adapted to supervised tasks and outperforms state-of …
Cited by:	6
Resources:	[PDF] from amazonaws.com

:view less

Determination of magnetopause and bow shock shape based on convolutional neural network modelling of MESSENGER data

2021

Alexander Lavrukhin, David Parunakian, Dmitry Nevsky, Sahib Julka, Michael Granitzer, A…
European Planetary Science Congress, EPSC2021-651
:read more

Authors:	Alexander Lavrukhin, David Parunakian, Dmitry Nevsky, Sahib Julka, Michael Granitzer, Andreas Windisch, Ute Amerstorfer
Publication date:	09/01/2021
Journal:	European Planetary Science Congress
Pages:	EPSC2021-651
Description:	The magnetosphere of Mercury is relatively small and highly dynamic, mostly due to the weak planetary magnetic field. Varying solar wind conditions principally determine the location of both the Hermean bow shock and magnetopause. In 2011-2015 MESSENGER spacecraft completed over 4000 orbits around Mercury, thus giving a data of more than 8000 crossings of bow shock and magnetopause of the planet, this makes it possible to study in detail the bow shock, the magnetopause and the magnetosheath structures. In this work we determine crossings of the bow shock and the magnetopause of Mercury by applying machine learning methods to the MESSENGER magnetometer data. We attempt to identify the crossings during the whole duration of the orbital mission and model the average three-dimensional shapes of these boundaries. The results are compared with those previously obtained in other works …

:view less

HateBERT: Retraining BERT for abusive language detection in English

2021

Tommaso Caselli, Valerio Basile, Jelena Mitrović, Michael Granitzer
Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021), 17-25
:read more

Authors:	Tommaso Caselli, Valerio Basile, Jelena Mitrović, Michael Granitzer
Publication date:	08/01/2021
Journal:	Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021)
Pages:	17-25
Description:	We introduce HateBERT, a re-trained BERT model for abusive language detection in English. The model was trained on RAL-E, a large-scale dataset of Reddit comments in English from communities banned for being offensive, abusive, or hateful that we have curated and made available to the public. We present the results of a detailed comparison between a general pre-trained language model and the retrained version on three English datasets for offensive, abusive language and hate speech detection tasks. In all datasets, HateBERT outperforms the corresponding general BERT model. We also discuss a battery of experiments comparing the portability of the fine-tuned models across the datasets, suggesting that portability is affected by compatibility of the annotated phenomena.
Cited by:	738
Resources:	[PDF] from aclanthology.org

:view less

Spatio-temporal machine learning analysis of social media data and refugee movement statistics

2021

Clemens Havas, Lorenz Wendlinger, Julian Stier, Sahib Julka, Veronika Krieger, Cornelia…
ISPRS International Journal of Geo-Information 10 (8), 498
:read more

Authors:	Clemens Havas, Lorenz Wendlinger, Julian Stier, Sahib Julka, Veronika Krieger, Cornelia Ferner, Andreas Petutschnig, Michael Granitzer, Stefan Wegenkittl, Bernd Resch
Publication date:	07/23/2021
Journal:	ISPRS International Journal of Geo-Information
Volume:	10
Issue:	8
Pages:	498
Publisher:	MDPI
Description:	In 2015, within the timespan of only a few months, more than a million people made their way from Turkey to Central Europe in the wake of the Syrian civil war. At the time, public authorities and relief organisations struggled with the admission, transfer, care, and accommodation of refugees due to the information gap about ongoing refugee movements. Therefore, we propose an approach utilising machine learning methods and publicly available data to provide more information about refugee movements. The approach combines methods to analyse the textual, temporal and spatial features of social media data and the number of arriving refugees of historical refugee movement statistics to provide relevant and up to date information about refugee movements and expected numbers. The results include spatial patterns and factual information about collective refugee movements extracted from social media data that match actual movement patterns. Furthermore, our approach enables us to forecast and simulate refugee movements to forecast an increase or decrease in the number of incoming refugees and to analyse potential future scenarios. We demonstrate that the approach proposed in this article benefits refugee management and vastly improves the status quo.
Cited by:	21
Resources:	[HTML] from mdpi.com

:view less

Design and Implementation of German Legal Decision Corpora.

2021

Stefanie Urchs, Jelena Mitrovic, Michael Granitzer
ICAART (2), 515-521
:read more

Authors:	Stefanie Urchs, Jelena Mitrovic, Michael Granitzer
Publication date:	07/01/2021
Journal:	ICAART (2)
Pages:	515-521
Description:	Law professionals are wordsmiths, their main tool is language. Therefore, the field of law produces a vast amount of written text. These texts have to be analysed, summarised, and used in the creation of new text, which is a task that reaches the limits of what is humanly possible. However, it is possible to automate this analysis by using Natural Language Processing techniques. To perform these techniques (annotated) text corpora are required. Unfortunately, publicly available (annotated) legal text corpora are rare. Even scarcer is the availability of (annotated) German legal text corpora. To meet this need for publicly available German legal text corpora this paper presents two German legal text corpora. The first corpus contains 32,748 decisions from 131 German courts, enriched with metadata. The second one is a subset of the first corpus and consists of 200 randomly chosen judgements. In these judgements a legal expert annotated the components conclusion, definition and subsumption of the German legal writing style Urteilsstil. Furthermore, the paper presents experiments on these corpora.
Cited by:	27
Resources:	[PDF] from scitepress.org

:view less

Automated identification of bias inducing words in news articles using linguistic and context-oriented features

2021

Timo Spinde, Lada Rudnitckaia, Jelena Mitrović, Felix Hamborg, Michael Granitzer, Bela …
Information Processing & Management 58 (3), 102505
:read more

Authors:	Timo Spinde, Lada Rudnitckaia, Jelena Mitrović, Felix Hamborg, Michael Granitzer, Bela Gipp, Karsten Donnay
Publication date:	05/01/2021
Journal:	Information Processing & Management
Volume:	58
Issue:	3
Pages:	102505
Publisher:	Pergamon
Description:	Media has a substantial impact on public perception of events, and, accordingly, the way media presents events can potentially alter the beliefs and views of the public. One of the ways in which bias in news articles can be introduced is by altering word choice. Such a form of bias is very challenging to identify automatically due to the high context-dependence and the lack of a large-scale gold-standard data set. In this paper, we present a prototypical yet robust and diverse data set for media bias research. It consists of 1,700 statements representing various media bias instances and contains labels for media bias identification on the word and sentence level. In contrast to existing research, our data incorporate background information on the participants’ demographics, political ideology, and their opinion about media in general. Based on our data, we also present a way to detect bias-inducing words in news articles …
Cited by:	140
Resources:	[HTML] from sciencedirect.com

:view less

DeepGG: a deep graph generator

2021

Julian Stier, Michael Granitzer
International Symposium on Intelligent Data Analysis, 313-324
:read more

Authors:	Julian Stier, Michael Granitzer
Publication date:	04/13/2021
Journal:	International Symposium on Intelligent Data Analysis
Pages:	313-324
Publisher:	Springer International Publishing
Description:	Learning distributions of graphs can be used for automatic drug discovery, molecular design, complex network analysis, and much more. We present an improved framework for learning generative models of graphs based on the idea of deep state machines. To learn state transition decisions we use a set of graph and node embedding techniques as memory of the state machine. Our analysis is based on learning the distribution of random graph generators for which we provide statistical tests to determine which properties can be learned and how well the original distribution of graphs is represented. We show that the design of the state machine favors specific distributions. Models of graphs of size up to 150 vertices are learned. Code and parameters are publicly available to reproduce our results.
Cited by:	7
Resources:	[PDF] from arxiv.org

:view less

Generative adversarial networks for automatic detection of mounds in digital terrain models (mars arabia terra)

2021

Sahib Julka, Michael Granitzer, Barbara De Toffoli, Luca Penasa, Riccardo Pozzobon, Ute…
EGU General Assembly Conference Abstracts, EGU21-9188
:read more

Authors:	Sahib Julka, Michael Granitzer, Barbara De Toffoli, Luca Penasa, Riccardo Pozzobon, Ute Amerstorfer
Publication date:	04/01/2021
Journal:	EGU General Assembly Conference Abstracts
Pages:	EGU21-9188
Description:	Mounds are positive relief features that can be ascribed to a variety of phenomena; they can be related to monogenic edifices due to spring or mud volcanism, rootless cones on top of lava flows, pingos and so on. In the case of sedimentary or spring case of mud extrusion, these mounds can be widespread regionally and/or contained in large complex craters, often in populations of several hundreds or thousands. Previous work on detection of such mounds in the Mars Arabia Terra involved exploiting morphometric parameters and mapping them onto Digital Terrain Models. In this work, we take a step further and develop more general methods to automatically detect them without explicitly defining the topographical features. We achieve this by using a generative framework trained in an adversarial fashion to produce realistic mappings with only a small number of training samples. Further, we introduce a terrain …
Cited by:	2

:view less

Evofficient: reproducing a cartesian genetic programming method

2021

Lorenz Wendlinger, Julian Stier, Michael Granitzer
European Conference on Genetic Programming (Part of EvoStar), 162-178
:read more

Authors:	Lorenz Wendlinger, Julian Stier, Michael Granitzer
Publication date:	03/25/2021
Journal:	European Conference on Genetic Programming (Part of EvoStar)
Pages:	162-178
Publisher:	Springer International Publishing
Description:	Designing Neural Network Architectures requires expert knowledge and extensive parameter searches. Neural Architecture Search (NAS) aims to change that by automating the design process. It is important that these approaches are reproducible so they can be used in real-life scenarios. In our work, we reproduce a genetic programming approach to designing convolutional neural networks called CGP-CNN. We show that this is difficult and requires many changes to the training scheme, reducing real-life applicability. We achieve a final accuracy of , substantially lower than the reported . This negates some of the benefits of using CGP-CNN for NAS. We establish a random search as a consensus baseline and show that it produces similar results to the evolutionary method of CGP-CNN. To assess the adaptability and generality of the presented algorithm, it is applied to CIFAR-100 and …
Cited by:	5

:view less

Automatic detection of magnetopause and bow shock crossing signatures in MESSENGER magnetometer data using Convolutional Neural Networks.

2021

Alexander Lavrukhin, David Parunakian, Dmitry Nevskiy, Sahib Julka, Michael Granitzer, …
EGU General Assembly Conference Abstracts, EGU21-12703
:read more

Authors:	Alexander Lavrukhin, David Parunakian, Dmitry Nevskiy, Sahib Julka, Michael Granitzer, Andreas Windisch, Christian Möstl, Martin A Reiss, Rachel L Bailey, Ute Amerstorfer
Publication date:	03/03/2021
Journal:	EGU General Assembly Conference Abstracts
Pages:	EGU21-12703
Description:	The magnetosphere of Mercury is rather small and highly dynamic, due to its weak internal magnetic field and its close proximity to the Sun. The changing solar wind conditions principally determine the locations of both the Hermean bow shock and magnetopause. In 2011–2015, the MESSENGER spacecraft completed more than 4000 orbits around Mercury, thus giving data of more than 8000 crossings of bow shock and magnetopause of the planet. This makes it possible to study in detail the bow shock, the magnetopause and the magnetosheath structures. In this work, we determine crossings of the bow shock and the magnetopause of Mercury by applying machine learning methods to the MESSENGER magnetometer data. We try to identify the crossings for the complete orbital mission and model the average three-dimensional shape of these boundaries depending on the external interplanetary magnetic field (IMF). Further, we try to clarify the dependence of the two boundary locations on the heliocentric distance of Mercury and on the solar activity cycle phase. Also, we study the effect of the IMF partial penetration into the Hermean magnetosphere. The results are compared with the obtained previously in other works.
Resources:	[PDF] from copernicus.org

:view less

Design and Implementation of German Legal Decision Corpora

2021

Michael Granitzer, Jelena Mitrović
Proceedings of the 13th International Conference on Agents and Artificial Intelligence
:read more

Authors:	Michael Granitzer, Jelena Mitrović
Publication date:	01/01/2021
Journal:	Proceedings of the 13th International Conference on Agents and Artificial Intelligence
Description:	Law professionals are wordsmiths, their main tool is language. Therefore, the field of law produces a vast amount of written text. These texts have to be analysed, summarised, and used in the creation of new text, which is a task that reaches the limits of what is humanly possible. However, it is possible to automate this analysis by using Natural Language Processing techniques. To perform these techniques (annotated) text corpora are required. Unfortunately, publicly available (annotated) legal text corpora are rare. Even scarcer is the availability of (annotated) German legal text corpora. To meet this need for publicly available German legal text corpora this paper presents two German legal text corpora. The first corpus contains 32,748 decisions from 131 German courts, enriched with metadata. The second one is a subset of the first corpus and consists of 200 randomly chosen judgements. In these judgements a legal expert annotated the components conclusion, definition and subsumption of the German legal writing style Urteilsstil. Furthermore, the paper presents experiments on these corpora.
Resources:	[HTML] from academia.edu

:view less

Analysis of a German Legal Citation Network.

2021

Tobias Milz, Michael Granitzer, Jelena Mitrovic
KDIR, 147-154
:read more

Authors:	Tobias Milz, Michael Granitzer, Jelena Mitrovic
Publication date:	01/01/2021
Journal:	KDIR
Pages:	147-154
Description:	The paper introduces the creation and analysis of a German legal citation network. The network consists of over 200.000 German court cases from all levels of appeal and jurisdiction and more than 50.000 laws. References to court decisions and laws are extracted from within the decision text of the court cases and added as links to the network. We apply network-based analysis techniques to support common legal information retrieval tasks such as identification of important court decisions and laws and case similarity searches. Furthermore, we demonstrate that the German case citation network displays scale-free behaviour, similar to that of the US and Austrian Supreme Courts as shown by previous research.
Cited by:	15
Resources:	[PDF] from scitepress.org

:view less

Correlation analysis between the robustness of sparse neural networks and their random hidden structural priors

2021

Mehdi Ben Amor, Julian Stier, Michael Granitzer
Procedia Computer Science 192, 4073-4082
:read more

Authors:	Mehdi Ben Amor, Julian Stier, Michael Granitzer
Publication date:	01/01/2021
Journal:	Procedia Computer Science
Volume:	192
Pages:	4073-4082
Publisher:	Elsevier
Description:	Deep learning models have been shown to be vulnerable to adversarial attacks. This perception led to analyzing deep learning models not only from the perspective of their performance measures but also their robustness to certain types of adversarial attacks. We take another step forward in relating the architectural structure of neural networks from a graph theoretic perspective to their robustness. We aim to investigate any existing correlations between graph theoretic properties and robustness of Sparse Neural Networks. Our hypothesis is, that graph theoretic properties as a prior of neural network structures is related to their robustness. To give answer to this hypothesis, we designed an empirical study with neural network models obtained through random graphs used as sparse structural priors for the networks. We additionally investigated the evaluation of a randomly pruned fully connected network as point of …
Cited by:	4
Resources:	[PDF] from sciencedirect.com

:view less

Multi-task network embedding with adaptive loss weighting

2020

Fatemeh Salehi Rizi, Michael Granitzer
2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mini…
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer
Publication date:	12/07/2020
Journal:	2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)
Pages:	1-5
Publisher:	IEEE
Description:	Network embedding is to learn low-dimensional representations of nodes which mostly preserve the network topological structure. In real-world networks, however, nodes are often associated with a rich set of attributes and labels which are potentially valuable in seeking more effective vector representations. To properly utilize this information, we propose a Joint Autoencoders framework for Multi-task network Embedding (JAME), which aims to encode a shared representation of local network structure, node attributes, and available node labels. Jointly embedding via multi-task learning is strongly dependent on the relative weighting between each task's loss function. Tuning these weights by hand is an expensive and difficult process, making multi-task learning prohibitive in practice. Therefore, we define an adaptive loss weighting layer capable of learning an optimal combination of loss weights during model …
Cited by:	6
Resources:	[PDF] from openreview.net

:view less

NLP_Passau at SemEval-2020 Task 12: Multilingual neural network for offensive language detection in English, Danish and Turkish

2020

Omar Hussein, Hachem Sfar, Jelena Mitrović, Michael Granitzer
Proceedings of the Fourteenth Workshop on Semantic Evaluation, 2090-2097
:read more

Authors:	Omar Hussein, Hachem Sfar, Jelena Mitrović, Michael Granitzer
Publication date:	12/01/2020
Journal:	Proceedings of the Fourteenth Workshop on Semantic Evaluation
Pages:	2090-2097
Description:	This paper describes a neural network (NN) model that was used for participating in the OffensEval, Task 12 of the SemEval 2020 workshop. The aim of this task is to identify offensive speech in social media, particularly in tweets. The model we used, C-BiGRU, is composed of a Convolutional Neural Network (CNN) along with a bidirectional Recurrent Neural Network (RNN). A multidimensional numerical representation (embedding) for each of the words in the tweets that were used by the model were determined using fastText. This allowed for using a dataset of labeled tweets to train the model on detecting combinations of words that may convey an offensive meaning. This model was used in the sub-task A of the English, Turkish and Danish competitions of the workshop, achieving F1 scores of 90.88%, 76.76% and 76.70%, respectively.
Cited by:	10
Resources:	[PDF] from aclanthology.org

:view less

nlpUP at SemEval-2020 Task 12: A blazing fast system for offensive language detection

2020

Ehab Hamdy, Jelena Mitrović, Michael Granitzer
Proceedings of the Fourteenth Workshop on Semantic Evaluation, 2098-2104
:read more

Authors:	Ehab Hamdy, Jelena Mitrović, Michael Granitzer
Publication date:	12/01/2020
Journal:	Proceedings of the Fourteenth Workshop on Semantic Evaluation
Pages:	2098-2104
Description:	In this paper, we introduce our submission for the SemEval Task 12, sub-tasks A and B for offensive language identification and categorization in English tweets. This year the data set for Task A is significantly larger than in the previous year. Therefore, we have adapted the BlazingText algorithm to extract embedding representation and classify texts after filtering and sanitizing the dataset according to the conventional text patterns on social media. We have gained both advantages of a speedy training process and obtained a good F1 score of 90.88% on the test set. For sub-task B, we opted to fine-tune a Bidirectional Encoder Representation from a Transformer (BERT) to accommodate the limited data for categorizing offensive tweets. We have achieved an F1 score of only 56.86%, but after experimenting with various label assignment thresholds in the pre-processing steps, the F1 score improved to 64%.
Cited by:	4
Resources:	[PDF] from aclanthology.org

:view less

Grupato at semeval-2020 task 12: Retraining mbert on social media and fine-tuned offensive language models

2020

Davide Colla, Tommaso Caselli, Valerio Basile, Jelena Mitrović, Michael Granitzer
Proceedings of the fourteenth workshop on semantic evaluation, 1546-1554
:read more

Authors:	Davide Colla, Tommaso Caselli, Valerio Basile, Jelena Mitrović, Michael Granitzer
Publication date:	12/01/2020
Journal:	Proceedings of the fourteenth workshop on semantic evaluation
Pages:	1546-1554
Description:	We introduce an approach to multilingual Offensive Language Detection based on the mBERT transformer model. We download extra training data from Twitter in English, Danish, and Turkish, and use it to re-train the model. We then fine-tuned the model on the provided training data and, in some configurations, implement transfer learning approach exploiting the typological relatedness between English and Danish. Our systems obtained good results across the three languages (. 9036 for EN,. 7619 for DA, and. 7789 for TR).
Cited by:	13
Resources:	[PDF] from aclanthology.org

:view less

Towards classifying parts of german legal writing styles in german legal judgments

2020

Stefanie Urchs, Jelena Mitrović, Michael Granitzer
2020 10th International Conference on Advanced Computer Information Technologies (ACIT)…
:read more

Authors:	Stefanie Urchs, Jelena Mitrović, Michael Granitzer
Publication date:	09/16/2020
Journal:	2020 10th International Conference on Advanced Computer Information Technologies (ACIT)
Pages:	451-454
Publisher:	IEEE
Description:	The main tool of a lawyer is their language. Legal prose is bound by writing styles, especially in Germany. These styles ensure that, i.a. judgments are written in a structured and comprehensive way. The writing style used for German judgements is called Urteilsstil and consist of several subcomponents. These subcomponents should be classifiable with the help of argumentation mining techniques. However, this classification is currently not possible because an annotated corpus, that considers such special structure of German legal text, is not available. This paper explores possibilities for classifying two subcomponents of the Urteilsstil by utilising argumentation mining. Furthermore, the creation of a new corpus for legal classification is proposed.
Cited by:	16
Resources:	[PDF] from github.io

:view less

Automatic detection of magnetopause and bow shock crossing signatures in MESSENGER magnetometer data

2020

Alexander Lavrukhin, David Parunakian, Dmitry Nevskiy, Ute Amerstorfer, Andreas Windisc…
European Planetary Science Congress, EPSC2020-826
:read more

Authors:	Alexander Lavrukhin, David Parunakian, Dmitry Nevskiy, Ute Amerstorfer, Andreas Windisch, Sahib Julka, Christian Möstl, Martin Reiss, Rachel Bailey
Publication date:	09/01/2020
Journal:	European Planetary Science Congress
Pages:	EPSC2020-826
Description:	The magnetosphere of Mercury is rather small and highly dynamic, due to its weak internal magnetic field and its close proximity to the Sun. The changing solar wind conditions principally determine the locations of both the Hermean bow shock and magnetopause. In 2011-2015 MESSENGER spacecraft completed more than 4000 orbits around Mercury, thus giving a data of more than 8000 crossings of bow shock and magnetopause of the planet. This makes it possible to study in detail the bow shock, the magnetopause and the magnetosheath structures. In this work, we determine crossings of the bow shock and the magnetopause of Mercury by applying machine learning methods to the MESSENGER magnetometer data. We try to identify the crossings for the complete orbital mission and model the average three-dimensional shape of these boundaries depending on the external interplanetary magnetic field …

:view less

Adsage: Anomaly detection in sequences of attributed graph edges applied to insider threat detection at fine-grained level

2020

Mathieu Garchery, Michael Granitzer
arXiv preprint arXiv:2007.06985
:read more

Authors:	Mathieu Garchery, Michael Granitzer
Publication date:	07/14/2020
Journal:	arXiv preprint arXiv:2007.06985
Description:	Previous works on the CERT insider threat detection case have neglected graph and text features despite their relevance to describe user behavior. Additionally, existing systems heavily rely on feature engineering and audit data aggregation to detect malicious activities. This is time consuming, requires expert knowledge and prevents tracing back alerts to precise user actions. To address these issues we introduce ADSAGE to detect anomalies in audit log events modeled as graph edges. Our general method is the first to perform anomaly detection at edge level while supporting both edge sequences and attributes, which can be numeric, categorical or even text. We describe how ADSAGE can be used for fine-grained, event level insider threat detection in different audit logs from the CERT use case. Remarking that there is no standard benchmark for the CERT problem, we use a previously proposed evaluation setting based on realistic recall-based metrics. We evaluate ADSAGE on authentication, email traffic and web browsing logs from the CERT insider threat datasets, as well as on real-world authentication events. ADSAGE is effective to detect anomalies in authentications, modeled as user to computer interactions, and in email communications. Simple baselines give surprisingly strong results as well. We also report performance split by malicious scenarios present in the CERT datasets: interestingly, several detectors are complementary and could be combined to improve detection. Overall, our results show that graph features are informative to characterize malicious insider activities, and that detection at fine-grained level is possible.
Cited by:	27
Resources:	[PDF] from arxiv.org

:view less

Improving intrusion detection systems using zero-shot recognition via graph embeddings

2020

Saber Zerhoudi, Michael Granitzer, Mathieu Garchery
2020 IEEE 44th Annual Computers, Software, and Applications Conference (COMPSAC), 790-797
:read more

Authors:	Saber Zerhoudi, Michael Granitzer, Mathieu Garchery
Publication date:	07/13/2020
Journal:	2020 IEEE 44th Annual Computers, Software, and Applications Conference (COMPSAC)
Pages:	790-797
Publisher:	IEEE
Description:	In order to detect insider threats, anomaly-based intrusion detection must learn profiles of normal user behavior. However this is particularly difficult when historical audit data is scarce. Zero-shot learning can address this limitation by compensating the absence of examples with semantic knowledge, allowing to better estimate behavior of unknown users. In this paper, we address insider threat detection in two use cases where historical user data is unavailable or obsolete. We extend an existing intrusion detection system by adding information describing user positions, roles and projects assignments. These semantic descriptions are encoded via graph embeddings. Experimental results show that providing this additional context improves insider threat detection significantly. This suggests that zero-shot learning is a promising way of improving intrusion detection systems.
Cited by:	22

:view less

Language proficiency scoring

2020

Cristina Arhiliuc, Jelena Mitrović, Michael Granitzer
Proceedings of the Twelfth Language Resources and Evaluation Conference, 5624-5630
:read more

Authors:	Cristina Arhiliuc, Jelena Mitrović, Michael Granitzer
Publication date:	05/01/2020
Journal:	Proceedings of the Twelfth Language Resources and Evaluation Conference
Pages:	5624-5630
Description:	The Common European Framework of Reference (CEFR) provides generic guidelines for the evaluation of language proficiency. Nevertheless, for automated proficiency classification systems, different approaches for different languages are proposed. Our paper evaluates and extends the results of an approach to Automatic Essay Scoring proposed as a part of the REPROLANG 2020 challenge. We provide a comparison between our results and the ones from the published paper and we include a new corpus for the English language for further experiments. Our results are lower than the expected ones when using the same approach and the system does not scale well with the added English corpus.
Cited by:	8
Resources:	[PDF] from aclanthology.org

:view less

I feel offended, don’t be abusive! implicit/explicit messages in offensive and abusive language

2020

Tommaso Caselli, Valerio Basile, Jelena Mitrović, Inga Kartoziya, Michael Granitzer
Proceedings of the twelfth language resources and evaluation conference, 6193-6202
:read more

Authors:	Tommaso Caselli, Valerio Basile, Jelena Mitrović, Inga Kartoziya, Michael Granitzer
Publication date:	05/01/2020
Journal:	Proceedings of the twelfth language resources and evaluation conference
Pages:	6193-6202
Description:	Abusive language detection is an unsolved and challenging problem for the NLP community. Recent literature suggests various approaches to distinguish between different language phenomena (eg, hate speech vs. cyberbullying vs. offensive language) and factors (degree of explicitness and target) that may help to classify different abusive language phenomena. There are data sets that annotate the target of abusive messages (ie OLID/OffensEval (Zampieri et al., 2019a)). However, there is a lack of data sets that take into account the degree of explicitness. In this paper, we propose annotation guidelines to distinguish between explicit and implicit abuse in English and apply them to OLID/OffensEval. The outcome is a newly created resource, AbuseEval v1. 0, which aims to address some of the existing issues in the annotation of offensive and abusive language (eg, explicitness of the message, presence of a target, need of context, and interaction across different phenomena).
Cited by:	260
Resources:	[PDF] from aclanthology.org

:view less

Influence of random walk parametrization on graph embeddings

2020

Fabian Schliski, Jörg Schlötterer, Michael Granitzer
European Conference on Information Retrieval, 58-65
:read more

Authors:	Fabian Schliski, Jörg Schlötterer, Michael Granitzer
Publication date:	04/08/2020
Journal:	European Conference on Information Retrieval
Pages:	58-65
Publisher:	Springer International Publishing
Description:	Network or graph embedding has gained increasing attention in the research community during the last years. In particular, many methods to create graph embeddings using random walk based approaches have been developed. node2vec introduced means to control the random walk behavior, guiding the walks. We aim to reproduce parts of their work and introduce two additional modifications (jump probabilities and attention to hubs), in order to investigate how guiding and modifying the walks influences the learned embeddings. The reproduction includes the case study illustrating homophily and structural equivalence subject to the chosen strategy and a node classification task. We were not able to illustrate structural equivalence and further results show that modifications of the walks only slightly improve node classification, if at all.
Cited by:	12
Resources:	[PDF] from nih.gov

:view less

QueryCrumbs search query history visualization–Usability, transparency and long-term usage

2020

Jörg Schlötterer, Christin Seifert, Christopher Satchell, Michael Granitzer
Journal of Computer Languages 57, 100941
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Christopher Satchell, Michael Granitzer
Publication date:	04/01/2020
Journal:	Journal of Computer Languages
Volume:	57
Pages:	100941
Publisher:	Elsevier
Description:	Models of human information seeking reveal that search, in particular ad-hoc retrieval, is non-linear and iterative. Despite these findings, today’s search user interfaces do not support non-linear navigation, like for example backtracking in time. We propose QueryCrumbs, a compact and easy-to-understand visualization for navigating the search query history supporting iterative query refinement. We apply a multi-layered interface design to support novices and first-time users as well as intermediate and expert users. The visualization is evaluated with novice users in a formative user study, with experts in a think aloud test and its usage in a long-term study with software logging. The formative evaluation showed that the interactions can be easily performed, and the visual encodings were well understood without instructions. Results indicate that QueryCrumbs can support users when searching for information in an …
Cited by:	2

:view less

Signed heterogeneous network embedding in social media

2020

Fatemeh Salehi Rizi, Michael Granitzer
Proceedings of the 35th Annual ACM Symposium on Applied Computing, 1877-1880
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer
Publication date:	03/30/2020
Journal:	Proceedings of the 35th Annual ACM Symposium on Applied Computing
Pages:	1877-1880
Description:	In today's social networks users can express positive or negative attitudes towards others, which forms sentiment links in addition to social links. This link heterogeneity builds a new network topology called signed heterogeneous networks. Meanwhile, network embeddings have emerged as convenient methods for conducting machine learning over networked data. Consequently, extending embedding methods for signed heterogeneous networks allows interesting analytical scenarios such as predicting the sign of sentiment links. Previous works mainly focus on heterogeneous network embedding discarding the sign information in the network. In this paper, we propose SiHet, a fast and scalable embedding method suited for Signed Heterogeneous networks. Our method extends classical network embedding algorithms through a new loss function and a sampling strategy which allows to integrate sign and social …
Cited by:	11
Resources:	[PDF] from acm.org

:view less

Shortest path distance approximation using deep learning techniques

2020

Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer
arXiv e-prints, arXiv: 2002.05257
:read more

Authors:	Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer
Publication date:	02/01/2020
Journal:	arXiv e-prints
Pages:	arXiv: 2002.05257
Description:	Computing shortest path distances between nodes lies at the heart of many graph algorithms and applications. Traditional exact methods such as breadth-first-search (BFS) do not scale up to contemporary, rapidly evolving today's massive networks. Therefore, it is required to find approximation methods to enable scalable graph processing with a significant speedup. In this paper, we utilize vector embeddings learnt by deep learning techniques to approximate the shortest paths distances in large graphs. We show that a feedforward neural network fed with embeddings can approximate distances with relatively low distortion error. The suggested method is evaluated on the Facebook, BlogCatalog, Youtube and Flickr social networks.
Cited by:	1

:view less

Cognitive Modeling in Computational Rhetoric: Litotes, Containment and the Unexcluded Middle.

2020

Jelena Mitrovic, Cliff O'Reilly, Randy Allen Harris, Michael Granitzer
ICAART (2), 806-813
:read more

Authors:	Jelena Mitrovic, Cliff O'Reilly, Randy Allen Harris, Michael Granitzer
Publication date:	02/01/2020
Journal:	ICAART (2)
Pages:	806-813
Description:	The focus of our study is the rhetorical figure litotes and its cognitive modeling. This figure is formed by a contrary term that emphatically accentuates a positive, eg He is not exactly an idiot (said of Albert Einstein). Lawrence Horn’s illumination of the Law of Excluded Middle and its relationship to litotes by creating an Unexcluded Middle is central to our ideas and we correlate this to Image Schema theories developed by Mark Johnson, George Lakoff and Rafael Núnez–specifically the schema of CONTAINMENT. The distinction between contrary and contradictory opposition is described. We extend the assessment of the Excluded Middle from the perspective of Image Schema theory into the realm of the Unexcluded Middle and draw a representation of the layout of containers and analogous concept-activation. Lastly, we create and present an OWL ontology and publish it online.
Cited by:	19
Resources:	[PDF] from scitepress.org

:view less

Global and Local Feature Learning for Ego-Network Analysis

2020

Fatemeh Salehi Rizi, Michael Granitzer, Konstantin Ziegler
arXiv e-prints, arXiv: 2002.06685
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer, Konstantin Ziegler
Publication date:	02/01/2020
Journal:	arXiv e-prints
Pages:	arXiv: 2002.06685
Description:	In an ego-network, an individual (ego) organizes its friends (alters) in different groups (social circles). This social network can be efficiently analyzed after learning representations of the ego and its alters in a low-dimensional, real vector space. These representations are then easily exploited via statistical models for tasks such as social circle detection and prediction. Recent advances in language modeling via deep learning have inspired new methods for learning network representations. These methods can capture the global structure of networks. In this paper, we evolve these techniques to also encode the local structure of neighborhoods. Therefore, our local representations capture network features that are hidden in the global representation of large networks. We show that the task of social circle prediction benefits from a combination of global and local features generated by our technique.

:view less

Abschlussbericht zum Teilvorhaben: Provenance Analytics für das Social Semantic Web der Dinge: im Verbundprojekt: Provenance Analytics: zur Pilotmaßnahme: Strukturwandel

2020

Michael Granitzer, Fatemeh Salehi Rizi, Torben Schnuchel
:read more

Authors:	Michael Granitzer, Fatemeh Salehi Rizi, Torben Schnuchel
Publication date:	01/01/2020
Publisher:	Universität Passau

:view less

Echtzeit-Lagebild für effizientes Migrationsmanagement zur Gewährleistung humanitärer Sicherheit (HUMAN+); Teilvorhaben: Integrative Echtzeit Lage-und Vorhersagemodelle für …

2020

Michael Granitzer, Sahib Julka, Julian Stier, Lorenz Wendlinger
:read more

Authors:	Michael Granitzer, Sahib Julka, Julian Stier, Lorenz Wendlinger
Publication date:	01/01/2020
Publisher:	Universität Passau

:view less

NAG: Neural Feature Aggregation Framework for Credit Card Fraud Detection

2020

M. Granitzer K. Ghosh Dastidar, J. Jurgovsky, W. Siblini, L. He-Guelton
2020 IEEE International Conference on Data Mining (ICDM), 92-101
:read more

Authors:	M. Granitzer K. Ghosh Dastidar, J. Jurgovsky, W. Siblini, L. He-Guelton
Publication date:	01/01/2020
Journal:	2020 IEEE International Conference on Data Mining (ICDM)
Pages:	92-101
Publisher:	10.1109/ICDM50108.2020.00018
Cited by:	44

:view less

Effects of Random Seeds on the Accuracy of Convolutional Neural Networks

2020

Michael Fellicious, Christofer and Weissgerber, Thomas and Granitzer
Machine Learning, Optimization, and Data Science, 93--102
:read more

Authors:	Michael Fellicious, Christofer and Weissgerber, Thomas and Granitzer
Publication date:	01/01/2020
Journal:	Machine Learning, Optimization, and Data Science
Pages:	93--102
Publisher:	Springer International Publishing
Cited by:	19

:view less

Towards automated feature engineering for credit card fraud detection using multi-perspective HMMs

2020

Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Liyun He-Guelton, Olivier Caelen, Mich…
Future Generation Computer Systems 102, 393-402
:read more

Authors:	Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Liyun He-Guelton, Olivier Caelen, Michael Granitzer, Sylvie Calabretto
Publication date:	01/01/2020
Journal:	Future Generation Computer Systems
Volume:	102
Pages:	393-402
Publisher:	North-Holland
Description:	Machine learning and data mining techniques have been used extensively in order to detect credit card frauds. However, most studies consider credit card transactions as isolated events and not as a sequence of transactions.In this framework, we model a sequence of credit card transactions from three different perspectives, namely (i) The sequence contains or doesn’t contain a fraud (ii) The sequence is obtained by fixing the card-holder or the payment terminal (iii) It is a sequence of spent amount or of elapsed time between the current and previous transactions. Combinations of the three binary perspectives give eight sets of sequences from the (training) set of transactions. Each one of these sequences is modelled with a Hidden Markov Model (HMM). Each HMM associates a likelihood to a transaction given its sequence of previous transactions. These likelihoods are used as additional features in a Random …
Cited by:	209
Resources:	[PDF] from sciencedirect.com

:view less

Competitive influence maximization: integrating budget allocation and seed selection

2019

Amirhossein Ansari, Masoud Dadgar, Ali Hamzeh, Jörg Schlötterer, Michael Granitzer
arXiv preprint arXiv:1912.12283
:read more

Authors:	Amirhossein Ansari, Masoud Dadgar, Ali Hamzeh, Jörg Schlötterer, Michael Granitzer
Publication date:	12/27/2019
Journal:	arXiv preprint arXiv:1912.12283
Description:	Today, many companies take advantage of viral marketing to promote their new products, and since there are several competing companies in many markets, Competitive Influence Maximization has attracted much attention. Two categories of studies exist in the literature. First, studies that focus on which nodes from the network to select considering the existence of the opponents. Second, studies that focus on the problem of budget allocation. In this work, we integrate these two lines of research and propose a new scenario where competition happens in two phases. First, parties identify the most influential nodes of the network. Then they compete over only these influential nodes by the amount of budget they allocate to each node. We provide a strong motivation for this integration. Also, in previous budget allocation studies, any fraction of the budget could be allocated to a node, which means that the action space was continuous. However, in our scenario, we consider that the action space is discrete. This consideration is far more realistic, and it significantly changes the problem. We model the scenario as a game and propose a novel framework for calculating the Nash equilibrium. Notably, building an efficient framework for this problem with such a huge action space and also handling the stochastic environment of influence maximization is very challenging. To tackle these difficulties, we devise a new payoff estimation method and a novel best response oracle to boost the efficiency of our framework. Also, we validate each aspect of our work through extensive experiments.
Cited by:	14
Resources:	[PDF] from arxiv.org

:view less

Parallel Total Variation Distance Estimation with Neural Networks for Merging Over-Clusterings

2019

Christian Reiser, Jörg Schlötterer, Michael Granitzer
arXiv preprint arXiv:1912.04022
:read more

Authors:	Christian Reiser, Jörg Schlötterer, Michael Granitzer
Publication date:	12/09/2019
Journal:	arXiv preprint arXiv:1912.04022
Description:	We consider the initial situation where a dataset has been over-partitioned into clusters and seek a domain independent way to merge those initial clusters. We identify the total variation distance (TVD) as suitable for this goal. By exploiting the relation of the TVD to the Bayes accuracy we show how neural networks can be used to estimate TVDs between all pairs of clusters in parallel. Crucially, the needed memory space is decreased by reducing the required number of output neurons from to . On realistically obtained over-clusterings of ImageNet subsets it is demonstrated that our TVD estimates lead to better merge decisions than those obtained by relying on state-of-the-art unsupervised representations. Further the generality of the approach is verified by evaluating it on a a point cloud dataset.
Resources:	[PDF] from arxiv.org

:view less

Credit card fraud detection using machine learning with integration of contextual knowledge

2019

Yvan Lucas
:read more

Authors:	Yvan Lucas
Publication date:	12/04/2019
Description:	The detection of credit card fraud has several features that make it a difficult task. First, attributes describing a transaction ignore sequential information. Secondly, purchasing behavior and fraud strategies can change over time, gradually making a decision function learned by an irrelevant classifier. We performed an exploratory analysis to quantify the day-by-day shift dataset and identified calendar periods that have different properties within the dataset. The main strategy for integrating sequential information is to create a set of attributes that are descriptive statistics obtained by aggregating cardholder transaction sequences. We used this method as a reference method for detecting credit card fraud. We have proposed a strategy for creating attributes based on Hidden Markov Models (HMMs) characterizing the transaction from different viewpoints in order to integrate a broad spectrum of sequential information within transactions. In fact, we model the authentic and fraudulent behaviors of merchants and cardholders according to two univariate characteristics: the date and the amount of transactions. Our multi-perspective approach based on HMM allows automated preprocessing of data to model temporal correlations. Experiments conducted on a large set of data from real-world credit card transactions (46 million transactions carried out by Belgian cardholders between March and May 2015) have shown that the proposed strategy for pre-processing data based on HMMs can detect more fraudulent transactions when combined with the Aggregate Data Pre-Processing strategy.
Cited by:	9
Resources:	[PDF] from hal.science

:view less

Infosky: visual exploration of large hierarchical document repositories

2019

Frank Kappe, Georg Drosch, Wolfgang Kienreich, Vedran Sabol, Jutta Becker, Keith Andrew…
Human-Centered Computing, 1268-1272
:read more

Authors:	Frank Kappe, Georg Drosch, Wolfgang Kienreich, Vedran Sabol, Jutta Becker, Keith Andrews, Michael Granitzer, Klaus Tochtermann, Peter Auer
Publication date:	11/11/2019
Journal:	Human-Centered Computing
Pages:	1268-1272
Publisher:	CRC Press
Description:	Abstract InfoSky is a system enabling users to explore large, hierarchically structured document collections. Similar to a real-world telescope, InfoSky employs a planar graphical representation with variable magnification. Documents are assumed to have significant textual content, which can be extracted if necessary with specialised tools. Documents of similar content are placed close to each other and are visualised as stars, forming clusters with distinct shapes. For greater performance, the hierarchical structure is exploited and force-directed placement is applied recursively at each level on much fewer objects, rather than on the whole corpus.
Cited by:	11

:view less

Policy learning for malaria control

2019

Van Bach Nguyen, Belaid Mohamed Karim, Bao Long Vu, Jörg Schlötterer, Michael Granitzer
arXiv preprint arXiv:1910.08926
:read more

Authors:	Van Bach Nguyen, Belaid Mohamed Karim, Bao Long Vu, Jörg Schlötterer, Michael Granitzer
Publication date:	10/20/2019
Journal:	arXiv preprint arXiv:1910.08926
Description:	Sequential decision making is a typical problem in reinforcement learning with plenty of algorithms to solve it. However, only a few of them can work effectively with a very small number of observations. In this report, we introduce the progress to learn the policy for Malaria Control as a Reinforcement Learning problem in the KDD Cup Challenge 2019 and propose diverse solutions to deal with the limited observations problem. We apply the Genetic Algorithm, Bayesian Optimization, Q-learning with sequence breaking to find the optimal policy for five years in a row with only 20 episodes/100 evaluations. We evaluate those algorithms and compare their performance with Random Search as a baseline. Among these algorithms, Q-Learning with sequence breaking has been submitted to the challenge and got ranked 7th in KDD Cup.
Cited by:	8
Resources:	[PDF] from arxiv.org

:view less

Inklusiv digital: Entwicklung eines Blended-Learning-Kurses für pädagogisch-pflegerische Fachkräfte

2019

Valerie Jochim, Susanne Eggert, Lisa Rußwurm, Thomas Weißgerber, Tom Knieper, Michael G…
merz| medien+ erziehung 63 (5), 32-39
:read more

Authors:	Valerie Jochim, Susanne Eggert, Lisa Rußwurm, Thomas Weißgerber, Tom Knieper, Michael Granitzer
Publication date:	09/25/2019
Journal:	merz\| medien+ erziehung
Volume:	63
Issue:	5
Pages:	32-39
Description:	Menschen mit sogenannter geistiger Behinderung sind im Umgang mit digitalen Medien auf Assistenz angewiesen. Die sie unterstützenden Fachkräfte sind auf diese Aufgabe oft nicht vorbereitet, da dies in der Ausbildung bisher kaum eine Rolle spielt. Inklusiv digital ist ein Blended-Learning-Kurs für pädagogisch-pflegerische Fachkräfte, dessen Online-Anteile über die Lernplattform Moodle realisiert werden. Ziel des Kurses ist es, Fachkräfte zu befähigen, digitale Medien in der Arbeit mit ihren Klientinnen und Klienten einzusetzen.
Cited by:	5
Resources:	[PDF] from merz-zeitschrift.de

:view less

An introduction to JRA4-Machine Learning in Europlanet-2024

2019

Tarek Al-Ubaidi, Roman Kern, Michael Granitzer, Maxim Khodachenko, Jörn Helbert, Nick C…
EPSC-DPS Joint Meeting 2019 2019, EPSC-DPS2019-1509
:read more

Authors:	Tarek Al-Ubaidi, Roman Kern, Michael Granitzer, Maxim Khodachenko, Jörn Helbert, Nick Cox, Jan Soucek, Sergio Fonte, Apostolos Christou, Igor Alexeev
Publication date:	09/01/2019
Journal:	EPSC-DPS Joint Meeting 2019
Volume:	2019
Pages:	EPSC-DPS2019-1509
Description:	Machine Learning is rapidly becoming a vital tool for data analysis and hence has great potential for applications in the course of scientific investigations in planetary science. A dedicated JRA will address a number of scientific challenges in planetary science that are expected to benefit significantly by the application of ML technologies with the aim of implementing a generic tool set, able to tackle a wide range of scientific problems with minimal customization effort. This presentation will give an overview of planned activities of EPN-2024-RI ML JRA4, its tasks and science cases and outline the main goals and envisioned ML tool development of the JRA.

:view less

Erweiterung datenbasierter Wertschöpfungsketten um transferierbare Modelle

2019

Torben Schnuchel, Michael Granitzer
Handbuch Industrie 4.0 und Digitale Transformation: Betriebswirtschaftliche, technische…
:read more

Authors:	Torben Schnuchel, Michael Granitzer
Publication date:	08/13/2019
Journal:	Handbuch Industrie 4.0 und Digitale Transformation: Betriebswirtschaftliche, technische und rechtliche Herausforderungen
Pages:	665-679
Publisher:	Springer Fachmedien Wiesbaden
Description:	Die Zukunftsvision der Fusion von industriellen Fertigungsprozessen und intelligenten cyberphysischen Systemen stellt Unternehmen vor Herausforderungen einer ganz besonderen Kategorie. Die Grundlage hierfür bildet der enorme technologische Fortschritt im Bereich der Datenanalyse in den letzten Jahren, der nun in unterschiedliche Bereiche unseres Lebens Einzug hält. Längst haben sich Privatpersonen damit arrangiert, neben ihrer eigentlichen Identität noch eine digitale Kopie von sich zu pflegen, die maschinenlesbar ist und somit Gegenstand von automatisierten Analysen (meist zu Werbezwecken) ist.
Cited by:	2

:view less

Investigating Extensions to Random Walk Based Graph Embedding

2019

Michael Schlötterer, Jörg and Wehking, Martin and Salehi Rizi, Fatemeh and Granitzer
IEEE International Conference on Cognitive Computing (ICCC), 81-90
:read more

Authors:	Michael Schlötterer, Jörg and Wehking, Martin and Salehi Rizi, Fatemeh and Granitzer
Publication date:	07/01/2019
Journal:	IEEE International Conference on Cognitive Computing (ICCC)
Pages:	81-90
Description:	Graph embedding has recently gained momentum in the research community, in particular after the introduction of random walk and neural network based approaches. However, most of the embedding approaches focus on representing the local neighborhood of nodes and fail to capture the global graph structure, i.e. to retain the relations to distant nodes. To counter that problem, we propose a novel extension to random walk based graph embedding, which removes a percentage of least frequent nodes from the walks at different levels. By this removal, we simulate farther distant nodes to reside in the close neighborhood of a node and hence explicitly represent their connection. Besides the common evaluation tasks for graph embeddings, such as node classification and link prediction, we evaluate and compare our approach against related methods on shortest path approximation. The results indicate, that …
Cited by:	24
Resources:	[PDF] from arxiv.org

:view less

Identifying and Clustering Users for Unsupervised Intrusion Detection in Corporate Audit Sessions

2019

Michael Garchery, Mathieu and Granitzer
IEEE International Conference on Cognitive Computing (ICCC), 19-28
:read more

Authors:	Michael Garchery, Mathieu and Granitzer
Publication date:	07/01/2019
Journal:	IEEE International Conference on Cognitive Computing (ICCC)
Pages:	19-28
Description:	We address intrusion detection in audit sessions, focusing on masquerades and insider threats. Unsupervised intrusion detection can straightforwardly be addressed through supervised user identification. This allows us to simply model the normal behavior of users implicitly within any supervised classifier. However certain users can have very similar behavior as shown by their audit sessions, thus learning to distinguish them is meaningless and leads to false positives. To address this issue we propose a second method, which identifies user clusters instead of individual users. By discarding harmless alarms for users with similar sessions, a better trade-off between false positives and detection rate can be achieved. We evaluate both methods on real-world and synthetic corporate audit sessions: our methods outperform anomaly detection baselines for masquerade detection. Our results suggest that user …
Cited by:	21

:view less

MMoveT15: A Twitter dataset for extracting and analysing migration-movement data of the European migration crisis 2015

2019

Stefanie Urchs, Lorenz Wendlinger, Jelena Mitrović, Michael Granitzer
2019 IEEE 28th International Conference on Enabling Technologies: Infrastructure for Co…
:read more

Authors:	Stefanie Urchs, Lorenz Wendlinger, Jelena Mitrović, Michael Granitzer
Publication date:	06/12/2019
Journal:	2019 IEEE 28th International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises (WETICE)
Pages:	146-149
Publisher:	IEEE
Description:	In the 2015 migration crisis thousands of refugees and migrants crossed the border to Hungary, Austria and Germany. The movements of these people are reflected in social media, especially on Twitter. In this paper we present a dataset of 3275 Tweets from the months September and October 2015. These Tweets are annotated regarding their relevance of containing quantitative movement information of refugees/migrants into Hungary, Austria and Germany. We present this dataset for a posterior analysis of the 2015 migration crisis or as a basis for creating an automated extraction / prediction system.
Cited by:	26
Resources:	[PDF] from researchgate.net

:view less

Dataset shift quantification for credit card fraud detection

2019

Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Sylvie Calabretto, Liyun He-Guelton, F…
2019 IEEE second international conference on artificial intelligence and knowledge engi…
:read more

Authors:	Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Sylvie Calabretto, Liyun He-Guelton, Frederic Oblé, Michael Granitzer
Publication date:	06/03/2019
Journal:	2019 IEEE second international conference on artificial intelligence and knowledge engineering (AIKE)
Pages:	97-100
Publisher:	IEEE
Description:	Machine learning and data mining techniques have been used extensively in order to detect credit card frauds. However purchase behaviour and fraudster strategies may change over time. This phenomenon is named dataset shift [1] or concept drift in the domain of fraud detection [2]. In this paper, we present a method to quantify day-by-day the dataset shift in our face-to-face credit card transactions dataset (card holder located in the shop) . In practice, we classify the days against each other and measure the efficiency of the classification. The more efficient the classification, the more different the buying behaviour between two days, and vice versa. Therefore, we obtain a distance matrix characterizing the dataset shift. After an agglomerative clustering of the distance matrix, we observe that the dataset shift pattern matches the calendar events for this time period (holidays, week-ends, etc). We then incorporate this …
Cited by:	46
Resources:	[PDF] from arxiv.org

:view less

nlpUP at SemEval-2019 task 6: A deep neural language model for offensive language detection

2019

Jelena Mitrović, Bastian Birkeneder, Michael Granitzer
Proceedings of the 13th international workshop on semantic evaluation, 722-726
:read more

Authors:	Jelena Mitrović, Bastian Birkeneder, Michael Granitzer
Publication date:	06/01/2019
Journal:	Proceedings of the 13th international workshop on semantic evaluation
Pages:	722-726
Description:	This paper presents our submission for the SemEval shared task 6, sub-task A on the identification of offensive language. Our proposed model, C-BiGRU, combines a Convolutional Neural Network (CNN) with a bidirectional Recurrent Neural Network (RNN). We utilize word2vec to capture the semantic similarities between words. This composition allows us to extract long term dependencies in tweets and distinguish between offensive and non-offensive tweets. In addition, we evaluate our approach on a different dataset and show that our model is capable of detecting online aggressiveness in both English and German tweets. Our model achieved a macro F1-score of 79.40% on the SemEval dataset.
Cited by:	42
Resources:	[PDF] from aclanthology.org

:view less

Mapping platforms into a new open science model for machine learning

2019

Thomas Weißgerber, Michael Granitzer
it - Information Technology - Methods and Applications of Informatics and Information T…
:read more

Authors:	Thomas Weißgerber, Michael Granitzer
Publication date:	04/26/2019
Journal:	it - Information Technology - Methods and Applications of Informatics and Information Technology
Publisher:	De Gruyter Online
Description:	Data-centric disciplines like machine learning and data science have become major research areas within computer science and beyond. However, the development of research processes and tools did not keep pace with the rapid advancement of the disciplines, resulting in several insufficiently tackled challenges to attain reproducibility, replicability, and comparability of achieved results. In this discussion paper, we review existing tools, platforms and standardization efforts for addressing these challenges. As a common ground for our analysis, we develop an open science centred process model for machine learning research, which combines openness and transparency with the core processes of machine learning and data science. Based on the features of over 40 tools, platforms and standards, we list the, in our opinion, 11 most central platforms for the research process in this paper. We conclude that most …
Cited by:	14

:view less

A comparitive evaluation of pruning techniques for Artificial Neural Networks

2019

Paul Häusner, P Häusner, M Granitzer
:read more

Authors:	Paul Häusner, P Häusner, M Granitzer
Publication date:	04/24/2019
Description:	Pruning of neural networks is one of several techniques that can be used to create sparse neural networks and to prevent neural networks from overfitting in general. In this thesis, different types of pruning techniques are compared with each other. Therefore, pruning techniques in general are divided into pruning strategies, which determine how many elements are deleted in each pruning step, and pruning methods that rank the weights based on their importance for the network. The different methods and strategies have been applied to some fully-connected networks that have previously been trained to a minimum on the MNIST dataset. Our experiments have shown that iterative pruning is especially useful if we want to create highly compressed networks, while fixed number pruning is able to create the highest performing ones. In contrast, single pruning is not able to create as good networks as the other two approaches do. From the different magnitude based methods that have been applied, the magnitude class blinded approach clearly outperforms the other ones. The Optimal Brain Damage method, which is based on the second order derivative of the cost function, performs nearly the same as the magnitude class blinded method, but with an increasing number of pruning steps the accuracy drops and the additional needed computations don’t pay off in our experiments. Furthermore, pruning in general is able to reduce the overfitting in the examined network dramatically and is even able to outperform other regularization techniques although more computational steps are required in the pruning approach compared to the other ones.
Cited by:	2
Resources:	[PDF] from github.io

:view less

Predicting event attendance exploring social influence

2019

Fatemeh Salehi Rizi, Michael Granitzer
Proceedings of the 34th ACM/SIGAPP symposium on applied computing, 2131-2134
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer
Publication date:	04/08/2019
Journal:	Proceedings of the 34th ACM/SIGAPP symposium on applied computing
Pages:	2131-2134
Description:	The problem of predicting people's participation in real-world events has received considerable attention as it offers valuable insights for human behavior analysis and event-related advertisement. Today social networks (e.g. Twitter) widely reflect large popular events where people discuss their interest with friends. Event participants usually stimulate friends to join the event which propagates a social influence in the network. In this paper, we propose to model the social influence of friends on event attendance. We consider non-geotagged posts besides structures of social groups to infer users' attendance. To leverage the information on network topology we apply some of recent graph embedding techniques such as node2vec, HARP and Poincarè. We describe the approach followed to design the feature space and feed it to a neural network. The performance evaluation is conducted using two large music …
Cited by:	16
Resources:	[PDF] from acm.org

:view less

Multiple perspectives HMM-based feature engineering for credit card fraud detection

2019

Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Sylvie Calabretto, Olivier Caelen, Liy…
Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing, 1359-1361
:read more

Authors:	Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Sylvie Calabretto, Olivier Caelen, Liyun He-Guelton, Michael Granitzer
Publication date:	04/08/2019
Journal:	Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing
Pages:	1359-1361
Description:	Machine learning and data mining techniques have been used extensively in order to detect credit card frauds. However, most studies consider credit card transactions as isolated events and not as a sequence of transactions. In this article, we model a sequence of credit card transactions from three different perspectives, namely (i) does the sequence contain a Fraud? (ii) Is the sequence obtained by fixing the card-holder or the payment terminal? (iii) Is it a sequence of spent amount or of elapsed time between the current and previous transactions? Combinations of the three binary perspectives give eight sets of sequences from the (training) set of transactions. Each one of these sets is modelled with a Hidden Markov Model (HMM). Each HMM associates a likelihood to a transaction given its sequence of previous transactions. These likelihoods are used as additional features in a Random Forest classifier for fraud …
Cited by:	63
Resources:	[PDF] from acm.org

:view less

Modeling legal terminology in sumo

2019

Jelena Mitrović
TOTh 2019, Terminology and Ontology: Theories and applications
:read more

Authors:	Jelena Mitrović
Publication date:	01/01/2019
Journal:	TOTh 2019, Terminology and Ontology: Theories and applications
Description:	We discuss ontological modeling of legal terminology in SUMO (Pease, 2001) in combination with the lexico-semantic database WordNet (Fellbaum, 1998). Formal systems that allow for automated semantic interpretation of law supported by lexical resources can provide solutions to many tasks related to legal reasoning. We wish to formalize legal issues in a computable language, using SUMO and its higher-order language to capture the semantics of the legal domain, which in itself relies on many quasi-logical rules. The Suggested Upper Merged Ontology (SUMO)(Niles and Pease, 2001) is a formal ontology stated in SUO-KIF (Pease 2003). SUMO began as an ontology of 1000 terms and 4000 axioms using those terms in SUO-KIF statements that describe the world. This set has since been expanded to more than 20,000 terms and 80,000 axioms. SUMO has been used in many academic and business applications, ranging from digital media (Pease&Rust, 2008), e-commerce, large scale theorem proving in first order (Pease et al, 2010) and higher order logic and natural language processing (Pease & Li, 2010). SUMO is a general purpose, upper level ontology but also a collection of domain specific ontologies, all of which are integrated and logically consistent with each other. Although SUMO is large, the scope of human knowledge is such that every domain cannot be covered in advance. WordNet is a general-purpose resource and similarly does not contain all specialized and technical terms used in English. While they have some basic legal notions, many more terms are needed in WordNet and in SUMO to cover common legal terms …
Cited by:	6
Resources:	[HTML] from academia.edu

:view less

Structural analysis of sparse neural networks

2019

Julian Stier, Michael Granitzer
Procedia Computer Science 159, 107-116
:read more

Authors:	Julian Stier, Michael Granitzer
Publication date:	01/01/2019
Journal:	Procedia Computer Science
Volume:	159
Pages:	107-116
Publisher:	Elsevier
Description:	Sparse Neural Networks regained attention due to their potential of mathematical and computational advantages. We give motivation to study Artificial Neural Networks (ANNs) from a network science perspective, provide a technique to embed arbitrary Directed Acyclic Graphs into ANNs and report study results on predicting the performance of image classifiers based on the structural properties of the networks’ underlying graph. Results could further progress neuroevolution and add explanations for the success of distinct architectures from a structural perspective.
Cited by:	24
Resources:	[PDF] from sciencedirect.com

:view less

Modeling Legal Terminology in SUMO

2019

Michael Granitzer Jelena Mitrović, Adam Pease
TOTh 2019, Terminology and Ontology: Theories and applications
:read more

Authors:	Michael Granitzer Jelena Mitrović, Adam Pease
Publication date:	01/01/2019
Journal:	TOTh 2019, Terminology and Ontology: Theories and applications

:view less

CNN features for Reverse Image Search

2018

Granitzer Michael Gaillard Mathieu, Egyed-Zsigmond Elöd
Document numérique 21 (1), 63-90
:read more

Authors:	Granitzer Michael Gaillard Mathieu, Egyed-Zsigmond Elöd
Publication date:	12/01/2018
Journal:	Document numérique
Volume:	21
Issue:	1
Pages:	63-90
Description:	Activations of units within top layers of convolutional neural networks (CNN features) are known to be good descriptors for image retrieval. In this paper, we investigate the use of CNN features for Reverse Image Search. We especially evaluate the robustness of these features against common modifications. In a first part, we present a benchmark to evaluate the retrieval performances of Reverse Image Search systems. To get a baseline, we evaluate well-established methods: Perceptual Hash Functions. In a second part, we further evaluate the retrieval performances of CNN features. We establish that CNN features perform better than Perceptual Hash Functions, even if neural networks are trained on unrelated classification tasks. CNN features are more robust against rotation and cropping. Finally, we give a list of layers from different neural networks that are remarkably good descriptors when used with a cosine …
Cited by:	7
Resources:	[PDF] from mgaillard.fr

:view less

Stacked gender prediction from tweet texts and images notebook for pan at CLEF 2018

2018

Giovanni Ciccone, Arthur Sultan, Léa Laporte, Elod Egyed-Zsigmond, Alaa Alhamzeh, Micha…
CLEF 2018-Conference and Labs of the Evaluation, 11p
:read more

Authors:	Giovanni Ciccone, Arthur Sultan, Léa Laporte, Elod Egyed-Zsigmond, Alaa Alhamzeh, Michael Granitzer
Publication date:	09/10/2018
Journal:	CLEF 2018-Conference and Labs of the Evaluation
Pages:	11p
Description:	This paper describes our participation at the PAN 2018 Author Profiling shared task. Given texts and images from some Twitter's authors, the goal is to estimate their genders. We considered all the languages (Arabic, English and Spanish) and all the prediction types (only from texts, only from images and combined). The final submitted system is a stacked classifier composed of two main parts. The first one, based on previous PAN Author Profiling editions, concerns gender prediction from texts. It consists in a pipeline of preprocessing, word n-grams from 1 to 2, TF-IDF with sublinear weighting, Linear Support Vector classification and probability calibration. The second part is formed by different layers of classifiers used for gender estimation from images: four base classifiers (object detection, face recognition, colour histograms, local binary patterns) in the first layer, a meta classifier in the second layer and an aggregation classifier as third layer. Finally, the two gender predictions, from texts and images, feed into the last layer classifier that provides the combined gender predictions.
Cited by:	15
Resources:	[PDF] from hal.science

:view less

Advanced Techniques for Signal Search and Automatic Classification of Observational Space Data

2018

Tarek Al-Ubaidi, Maxim Khodachenko, Roman Kern, Michael Granitzer, Stefaan Poedts
European Planetary Science Congress, EPSC2018-844
:read more

Authors:	Tarek Al-Ubaidi, Maxim Khodachenko, Roman Kern, Michael Granitzer, Stefaan Poedts
Publication date:	09/01/2018
Journal:	European Planetary Science Congress
Pages:	EPSC2018-844
Description:	An outline of various approaches in machine learning and content based search, investigated by members of the former IMPEx-FP7 project consortium, in close cooperation with partners Know-Center, Graz University of Technology, and University of Passau. Novel possibilities that open up, using these or equivalent techniques in conjunction with space science are presented. Using a concrete example based on prototype implementations, it is shown that applications of machine learning and content based search allow systems to effectively classify and pre-select scientific data and hence speed up scientific workflows significantly. The implications and possibilities for the emerging field of e-Science are discussed.

:view less

QueryCrumbs for Experts: A Compact Visual Query Support System to Facilitate Insights into Search Engine Internals

2018

Jörg Schlötterer, Christin Seifert, Michael Granitzer
2018 22nd International Conference Information Visualisation (IV), 78-84
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Michael Granitzer
Publication date:	07/10/2018
Journal:	2018 22nd International Conference Information Visualisation (IV)
Pages:	78-84
Publisher:	IEEE
Description:	Search experts use advanced query language and search tactics to formulate their queries. However, the effectiveness of those advanced techniques depends on the search engine internals. We propose QueryCrumbs for Experts, a compact visualization, which facilitates insights to the search engine internals and therefore allows the searcher to determine effective search strategies. Treating the search engine as a black box, QueryCrumbs can be seamlessly integrated into existing search interfaces, guiding the user's exploration and assessment of results. QueryCrumbs for Experts visualize the recent search history alongside with a simple and also a qualitative comparison of the result sets, from which conclusions about the search engine internals can be drawn. The evaluation shows that, by identifying specific patterns in the visualization, expert users can gain valuable insights into search engine internals …
Cited by:	2
Resources:	[PDF] from utwente.nl

:view less

All-implicants neural networks for efficient boolean function representation

2018

Federico Buffoni, Gabriele Gianini, Ernesto Damiani, Michael Granitzer
2018 IEEE International Conference on Cognitive Computing (ICCC), 82-86
:read more

Authors:	Federico Buffoni, Gabriele Gianini, Ernesto Damiani, Michael Granitzer
Publication date:	07/02/2018
Journal:	2018 IEEE International Conference on Cognitive Computing (ICCC)
Pages:	82-86
Publisher:	IEEE
Description:	Boolean classifiers can be evolved by means of genetic algorithms. This can be done within an intercommunicating island system, of evolutionary niches, undergoing cycles that alternate long periods of isolation to short periods of information exchange. In these settings, the efficiency of the communication is a key requirement. In the present work, we address this requirement by providing a technique for efficiently representing and transmitting differential encodings of Boolean functions. We introduce a new class of Boolean Neural Networks (BNN), the all-implicants BNN, and show that this representation supports efficient update communication, better than the classical representation, based on truth tables.
Cited by:	1

:view less

Sequence classification for credit-card fraud detection

2018

Johannes Jurgovsky, Michael Granitzer, Konstantin Ziegler, Sylvie Calabretto, Pierre-Ed…
Expert systems with applications 100, 234-245
:read more

Authors:	Johannes Jurgovsky, Michael Granitzer, Konstantin Ziegler, Sylvie Calabretto, Pierre-Edouard Portier, Liyun He-Guelton, Olivier Caelen
Publication date:	06/15/2018
Journal:	Expert systems with applications
Volume:	100
Pages:	234-245
Publisher:	Pergamon
Description:	Due to the growing volume of electronic payments, the monetary strain of credit-card fraud is turning into a substantial challenge for financial institutions and service providers, thus forcing them to continuously improve their fraud detection systems. However, modern data-driven and learning-based methods, despite their popularity in other domains, only slowly find their way into business applications.In this paper, we phrase the fraud detection problem as a sequence classification task and employ Long Short-Term Memory (LSTM) networks to incorporate transaction sequences. We also integrate state-of-the-art feature aggregation strategies and report our results by means of traditional retrieval metrics.A comparison to a baseline random forest (RF) classifier showed that the LSTM improves detection accuracy on offline transactions where the card-holder is physically present at a merchant. Both the sequential and …
Cited by:	865

:view less

Collection-Document Summaries

2018

Nils Witt, Michael Granitzer, Christin Seifert
European Conference on Information Retrieval, 638-643
:read more

Authors:	Nils Witt, Michael Granitzer, Christin Seifert
Publication date:	03/01/2018
Journal:	European Conference on Information Retrieval
Pages:	638-643
Publisher:	Springer International Publishing
Description:	Learning something new from a text requires the reader to build on existing knowledge and add new material at the same time. Therefore, we propose collection-document (CDS) summaries that highlight commonalities and differences between a collection (or a single document) and a single document. We devise evaluation metrics that do not require human judgement, and three algorithms for extracting CDS that are based on single-document keyword-extraction methods. Our evaluation shows that different algorithms have different strengths, e.g. TF-IDF based approach best describes document overlap while the adaption of Rake provides keywords with a broad topical coverage. The proposed criteria and procedure can be used to evaluate document-collection summaries without annotated corpora or provide additional insight in an evaluation with human-generated ground truth.
Resources:	[PDF] from utwente.nl

:view less

Shortest Path Distance Approximation Using Deep Learning Techniques

2018

Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer
2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mini…
:read more

Authors:	Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer
Publication date:	01/01/2018
Journal:	2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)
Publisher:	IEEE, 10.1109/ASONAM.2018.8508763
Description:	Computing shortest path distances between nodes lies at the heart of many graph algorithms and applications. Traditional exact methods such as breadth-first-search (BFS) do not scale up to contemporary, rapidly evolving today's massive networks. Therefore, it is required to find approximation methods to enable scalable graph processing with a significant speedup. In this paper, we utilize vector embeddings learnt by deep learning techniques to approximate the shortest paths distances in large graphs. We show that a feedforward neural network fed with embeddings can approximate distances with relatively low distortion error. The suggested method is evaluated on the Facebook, BlogCatalog, Youtube and Flickr social networks.
Cited by:	69
Resources:	[PDF] from arxiv.org

:view less

Analysing neural network topologies: a game theoretic approach

2018

Julian Stier, Gabriele Gianini, Michael Granitzer, Konstantin Ziegler
Procedia Computer Science 126, 234-243
:read more

Authors:	Julian Stier, Gabriele Gianini, Michael Granitzer, Konstantin Ziegler
Publication date:	01/01/2018
Journal:	Procedia Computer Science
Volume:	126
Pages:	234-243
Publisher:	Elsevier
Description:	Artificial Neural Networks have shown impressive success in very different application cases. Choosing a proper network architecture is a critical decision for a network’s success, usually done in a manual manner. As a straightforward strategy, large, mostly fully connected architectures are selected, thereby relying on a good optimization strategy to find proper weights while at the same time avoiding overfitting. However, large parts of the final network are redundant. In the best case, large parts of the network become simply irrelevant for later inferencing. In the worst case, highly parameterized architectures hinder proper optimization and allow the easy creation of adverserial examples fooling the network. A first step in removing irrelevant architectural parts lies in identifying those parts, which requires measuring the contribution of individual components such as neurons. In previous work, heuristics based on using …
Cited by:	40
Resources:	[PDF] from sciencedirect.com

:view less

On the influence of categorical features in ranking anomalies using mixed data

2018

Mathieu Garchery, Michael Granitzer
Procedia Computer Science 126, 77-86
:read more

Authors:	Mathieu Garchery, Michael Granitzer
Publication date:	01/01/2018
Journal:	Procedia Computer Science
Volume:	126
Pages:	77-86
Publisher:	Elsevier
Description:	Most unsupervised anomaly ranking approaches are compatible with numeric data only, leading to categorical features often being ignored in practice. Even though some methods address this issue, few support mixed data and the influence of excluding or including categorical attributes has not been studied well yet. In this paper, we take a first step towards considering categorical and numeric attributes jointly for unsupervised anomaly ranking by benchmarking selected methods. We introduce three new approaches: two entropy-based methods based on individual and collective entropy contribution, as well as an extension of Isolation Forest supporting mixed data, and benchmark them against SPAD, a state-of-the-art probabilistic anomaly ranker. We observe that our entropy methods detect very similar anomalies in practice, and these anomalies are mostly globally isolated observations. Both entropy methods …
Cited by:	17
Resources:	[PDF] from sciencedirect.com

:view less

Database and Expert Systems Applications

2018

Mourad Elloumi, Michael Granitzer, Abdelkader Hameurlain, Christin Seifert, Benno Stein…
:read more

Authors:	Mourad Elloumi, Michael Granitzer, Abdelkader Hameurlain, Christin Seifert, Benno Stein, A Min Tjoa, Roland Wagner
Publication date:	01/01/2018
Publisher:	Springer International Publishing
Description:	The Database and Expert Systems Applications (DEXA) workshops are a platform for the exchange of ideas, experiences, and opinions among theoreticians and practitioners–those who are defining requirements for future systems in the areas of database and artificial technologies.
Cited by:	1
Resources:	[HTML] from mehralborz.ac.ir

:view less

Properties of vector embeddings in social networks

2017

Fatemeh Salehi Rizi, Michael Granitzer
Algorithms 10 (4), 109
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer
Publication date:	09/27/2017
Journal:	Algorithms
Volume:	10
Issue:	4
Pages:	109
Publisher:	MDPI
Description:	Embedding social network data into a low-dimensional vector space has shown promising performance for many real-world applications, such as node classification, node clustering, link prediction and network visualization. However, the information contained in these vector embeddings remains abstract and hard to interpret. Methods for inspecting embeddings usually rely on visualization methods, which do not work on a larger scale and do not give concrete interpretations of vector embeddings in terms of preserved network properties (e.g., centrality or betweenness measures). In this paper, we study and investigate network properties preserved by recent random walk-based embedding procedures like node2vec, DeepWalk or LINE. We propose a method that applies learning to rank in order to relate embeddings to network centralities. We evaluate our approach with extensive experiments on real-world and artificial social networks. Experiments show that each embedding method learns different network properties. In addition, we show that our graph embeddings in combination with neural networks provide a computationally efficient way to approximate the Closeness Centrality measure in social networks.
Cited by:	40
Resources:	[HTML] from mdpi.com

:view less

TIR 2017. 14th international workshop on technologies for information retrieval: Preface

2017

Michael Granitzer, Christin Seifert, Benno Stein
28th International Conference on Database and Expert Systems Applications, DEXA 2017
:read more

Authors:	Michael Granitzer, Christin Seifert, Benno Stein
Publication date:	09/25/2017
Journal:	28th International Conference on Database and Expert Systems Applications, DEXA 2017
Publisher:	IEEE
Description:	Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

:view less

Global and local feature learning for ego-network analysis

2017

Fatemeh Salehi Rizi, Michael Granitzer, Konstantin Ziegler
2017 28th International Workshop on Database and Expert Systems Applications (DEXA), 98…
:read more

Authors:	Fatemeh Salehi Rizi, Michael Granitzer, Konstantin Ziegler
Publication date:	08/28/2017
Journal:	2017 28th International Workshop on Database and Expert Systems Applications (DEXA)
Pages:	98-102
Publisher:	IEEE
Description:	In an ego-network, an individual (ego) organizes its friends (alters) in different groups (social circles). This social network can be efficiently analyzed after learning representations of the ego and its alters in a low-dimensional, real vector space. These representations are then easily exploited via statistical models for tasks such as social circle detection and prediction. Recent advances in language modeling via deep learning have inspired new methods for learning network representations. These methods can capture the global structure of networks. In this paper, we evolve these techniques to also encode the local structure of neighborhoods. Therefore, our local representations capture network features that are hidden in the global representation of large networks. We show that the task of social circle prediction benefits from a combination of global and local features generated by our technique.
Cited by:	7
Resources:	[PDF] from arxiv.org

:view less

On joint representation learning of network structure and document content

2017

Jörg Schlötterer, Christin Seifert, Michael Granitzer
International Cross-Domain Conference for Machine Learning and Knowledge Extraction, 23…
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Michael Granitzer
Publication date:	08/24/2017
Journal:	International Cross-Domain Conference for Machine Learning and Knowledge Extraction
Pages:	237-251
Publisher:	Springer International Publishing
Description:	Inspired by the advancements of representation learning for natural language processing, learning continuous feature representations of nodes in networks has recently gained attention. Similar to word embeddings, node embeddings have been shown to capture certain semantics of the network structure. Combining both research directions into a joint representation learning of network structure and document content seems a promising direction to increase the quality of the learned representations. However, research is typically focused on either word or network embeddings and few approaches that learn a joint representation have been proposed. We present an overview of that field, starting at word representations, moving over document and network node representations to joint representations. We make the connections between the different models explicit and introduce a novel model for learning a joint …
Cited by:	7
Resources:	[PDF] from hal.science

:view less

Querycrumbs: A compact visualization for navigating the search query history

2017

Christin Seifert, Jörg Schlötterer, Michael Granitzer
2017 21st International Conference Information Visualisation (IV), 35-44
:read more

Authors:	Christin Seifert, Jörg Schlötterer, Michael Granitzer
Publication date:	07/11/2017
Journal:	2017 21st International Conference Information Visualisation (IV)
Pages:	35-44
Publisher:	IEEE
Description:	Models of human information seeking reveal that search, in particular ad-hoc retrieval, is non-linear and iterative. Despite these findings, todays search user interfaces do not support non-linear navigation, like for example backtracking in time. In this work, we propose QueryCrumbs, a compact and easy-to-understand visualization for navigating the search query history supporting iterative query refinement. We apply a multi-layered interface design to support novices and firsttime users as well as intermediate users. The formative evaluation with first-time and intermediate users showed that the interactions can be easily performed, and the visual encodings were well understood without instructions. Results indicate that QueryCrumbs can support users when searching for information in an iterative manner.
Cited by:	4
Resources:	[PDF] from github.io

:view less

Injecting semantic background knowledge into neural networks using graph embeddings

2017

Konstantin Ziegler, Olivier Caelen, Mathieu Garchery, Michael Granitzer, Liyun He-Guelt…
2017 IEEE 26th international conference on enabling technologies: Infrastructure for co…
:read more

Authors:	Konstantin Ziegler, Olivier Caelen, Mathieu Garchery, Michael Granitzer, Liyun He-Guelton, Johannes Jurgovsky, Pierre-Edouard Portier, Stefan Zwicklbauer
Publication date:	06/21/2017
Journal:	2017 IEEE 26th international conference on enabling technologies: Infrastructure for collaborative enterprises (WETICE)
Pages:	200-205
Publisher:	IEEE
Description:	The inferences of a machine learning algorithm are naturally limited by the available data. In many real-world applications, the provided internal data is domain-specific and we use external background knowledge to derive or add new features. Semantic networks, like linked open data, provide a largely unused treasure trove of background knowledge. This drives a recent surge of interest in unsupervised methods to automatically extract such semantic background knowledge and inject it into machine learning algorithms. In this work, we describe the general process of extracting knowledge from semantic networks through vector space embeddings. The locations in the vector space then reflect relations in the original semantic network. We perform this extraction for geographic background knowledge and inject it into a neural network for the complicated real-world task of credit-card fraud detection. This improves …
Cited by:	32
Resources:	[PDF] from academia.edu

:view less

Ubiquitous access to digital cultural heritage

2017

Christin Seifert, Werner Bailer, Thomas Orgel, Louis Gantner, Roman Kern, Hermann Ziak,…
Journal on Computing and Cultural Heritage (JOCCH) 10 (1), 1-27
:read more

Authors:	Christin Seifert, Werner Bailer, Thomas Orgel, Louis Gantner, Roman Kern, Hermann Ziak, Albin Petit, Jörg Schlötterer, Stefan Zwicklbauer, Michael Granitzer
Publication date:	04/14/2017
Journal:	Journal on Computing and Cultural Heritage (JOCCH)
Volume:	10
Issue:	1
Pages:	1-27
Publisher:	ACM
Description:	The digitization initiatives in the past decades have led to a tremendous increase in digitized objects in the cultural heritage domain. Although digitally available, these objects are often not easily accessible for interested users because of the distributed allocation of the content in different repositories and the variety in data structure and standards. When users search for cultural content, they first need to identify the specific repository and then need to know how to search within this platform (e.g., usage of specific vocabulary). The goal of the EEXCESS project is to design and implement an infrastructure that enables ubiquitous access to digital cultural heritage content. Cultural content should be made available in the channels that users habitually visit and be tailored to their current context without the need to manually search multiple portals or content repositories. To realize this goal, open-source software …
Cited by:	44
Resources:	[PDF] from acm.org

:view less

Discovering, ranking and merging rdf data cubes

2017

Sebastian Bayerl, Michael Granitzer
2017 IEEE 11th International Conference on Semantic Computing (ICSC), 133-140
:read more

Authors:	Sebastian Bayerl, Michael Granitzer
Publication date:	01/30/2017
Journal:	2017 IEEE 11th International Conference on Semantic Computing (ICSC)
Pages:	133-140
Publisher:	IEEE
Description:	The RDF Data Cube Vocabulary [1] is a W3C recommendation for publishing multi-dimensional data in a semantic web format. A large and growing number of such data cubes is already available in the linked data cloud. Merging multiple isolated cubes can lead to new insights as known from traditional data warehousing. In order to access and integrate the decentralized cubes, a suitable ranking and discovery mechanism is needed. We propose a cube discovery approach, based on the structure definition of the cubes. Syntactic and semantic properties of the cubes are considered to develop a similarity measure for cubes. We present a graph-based shortest-path algorithm that utilizes the DBpedia category dataset and a Word2Vec [2] model, to carry out the discovery process. The structural mapping for the data cubes generated during our ranking process can be leveraged to support the cube integration process …
Cited by:	5

:view less

Identifying Tweets from the Economic Domain

2017

Alexander Böhm, Christin Seifert, Jörg Schlötterer, Michael Granitzer
Theorie, Semantik und Organisation von Wissen, 161-166
:read more

Authors:	Alexander Böhm, Christin Seifert, Jörg Schlötterer, Michael Granitzer
Publication date:	01/18/2017
Journal:	Theorie, Semantik und Organisation von Wissen
Pages:	161-166
Publisher:	Ergon-Verlag
Description:	The rapid spread of information on the microblogging platform Twitter bears great potential for scientists to share their ideas and keep track on recent research results. However, Twitter lacks the ability to identify (and hence to connect to) other users in a particular domain. We adapt an approach to identify computer scientists on Twitter to the domain of economics. We learn a model, which classifies users as economist or non-economist, based on ground-truth data obtained by means of an initial set of Twitter accounts of economic journals. We further conduct an analysis of content-based and network features. Our evaluation shows, that economists can be detected rather accurately on twitter based on tweet terms (F1= 0, 7 with Naive Bayes). Neither adding hash-tags nor followers could further improve the classification model.
Cited by:	1

:view less

Kollaborative Unterstützung von Arbeits-und Lernprozessen im Unternehmen mit mobilen interaktiven Multimedia-Anwendungen: Schlussbericht: Laufzeit des Vorhabens: 01.09. 2013-31 …

2017

Michael Granitzer, Gerrit Hornung, Thomas Knieper, Harald Kosch, Franz Lehner, Kai Hofm…
:read more

Authors:	Michael Granitzer, Gerrit Hornung, Thomas Knieper, Harald Kosch, Franz Lehner, Kai Hofmann, Stefan John, Michael Langbauer, Katrin Tonndorf, Julian Windscheid
Publication date:	01/01/2017
Publisher:	Universität Passau

:view less

INSA LYON and UNI PASSAU’s Participation at PAN@ CLEF’17: Author Profiling task.

2017

Guillaume Kheng, Léa Laporte, Michael Granitzer
CLEF (Working notes)
:read more

Authors:	Guillaume Kheng, Léa Laporte, Michael Granitzer
Publication date:	01/01/2017
Journal:	CLEF (Working notes)
Description:	This paper describes the participation of INSA Lyon and UNI Passau at the PAN 2017 Author Profiling task. Given the language and tweets from an author, the goal is to predict his/her gender and language variety. We consider two strategies: a" loose" classification that learns one predictive model for the gender and another one for the variety, and a" successive" classification that first predict the gender then learn a predictive model for variety, given the gender. We consider all the languages. We experiment various features representations and machine learning algorithms used in previous PAN Author Profiling editions in order to learn the models. We adapt the features and machine learning algorithm used for each language and each classification task by selecting the configuration that provides the best results in terms of prediction performance.
Cited by:	8
Resources:	[PDF] from ceur-ws.org

:view less

On the quality of semantic interest profiles for onine social network consumers

2016

Christoph Besel, Jörg Schlötterer, Michael Granitzer
ACM SIGAPP Applied Computing Review 16 (3), 5-14
:read more

Authors:	Christoph Besel, Jörg Schlötterer, Michael Granitzer
Publication date:	11/04/2016
Journal:	ACM SIGAPP Applied Computing Review
Volume:	16
Issue:	3
Pages:	5-14
Publisher:	ACM
Description:	Social media based recommendation systems infer user' interests and preferences from their social network activity in order to provide personalised recommendations. Typically, the user profiles are generated by analysing the users' posts or tweets. However, there might be a significant difference between what a user produces and what she consumes. We propose an approach for inferring user interests from followees (the accounts the user follows) rather than tweets. This is done by extracting named entities from a user's followees using the English Wikipedia as knowledge base and regarding them as interests. Afterwards, a spreading activation algorithm is performed on a Wikipedia category taxonomy to aggregate the various interests to a more abstract and broader interest profile. We evaluate the coverage of followee lists in terms of named entities and show that they provide sufficient input to infer …
Cited by:	17
Resources:	[PDF] from acm.org

:view less

A dual screen concept for user-controlled hypervideo-based physiotherapy training

2016

Britta Meixner, Christian Handschigl, Stefan John, Michael Granitzer
Proceedings of the 2016 ACM Workshop on Multimedia for Personal Health and Health Care,…
:read more

Authors:	Britta Meixner, Christian Handschigl, Stefan John, Michael Granitzer
Publication date:	10/16/2016
Journal:	Proceedings of the 2016 ACM Workshop on Multimedia for Personal Health and Health Care
Pages:	9-13
Description:	Dual screen concepts for hypervideo-based physiotherapy training are important in healthcare settings, but existing applications often cannot be adapted to personal needs and do not support correct posture. In this paper, we describe the design and implementation of a dual screen application (handheld and TV) that allows patients to view hypervideos designed to help them correctly perform their exercises. This approach lets patients adapt their training to their daily needs and their overall training progress. We evaluated this prototypical implementation in a user test with post-operative care prostate cancer patients. From our results, we derived design recommendations for dual screen physical training hypervideo applications.
Cited by:	3
Resources:	[PDF] from acm.org

:view less

Second Screen Hypervideo-Based Physiotherapy Training: INVITED DEMO PAPER

2016

Christian Handschigl, Britta Meixner, Stefan John, Michael Granitzer
Proceedings of the 2016 ACM Workshop on Multimedia for Personal Health and Health Care,…
:read more

Authors:	Christian Handschigl, Britta Meixner, Stefan John, Michael Granitzer
Publication date:	10/16/2016
Journal:	Proceedings of the 2016 ACM Workshop on Multimedia for Personal Health and Health Care
Pages:	51-51
Description:	Adapting to personal needs and supporting correct posture are important in physiotherapy training. In this demo, we show a dual screen application (handheld and TV) that allows patients to view hypervideo training programs. Designed to guide their daily exercises, these programs can be adapted to daily needs. The dual screen concept offers the positional flexibility missing in single screen solutions.
Resources:	[PDF] from acm.org

:view less

Hypervideo Production Using Crowdsourced Youtube Videos

2016

Stefan John, Christian Handschigl, Britta Meixner, Michael Granitzer
Proceedings of the 24th ACM international conference on Multimedia, 695-697
:read more

Authors:	Stefan John, Christian Handschigl, Britta Meixner, Michael Granitzer
Publication date:	10/01/2016
Journal:	Proceedings of the 24th ACM international conference on Multimedia
Pages:	695-697
Description:	Hypervideos, consisting of media enriched and linked video scenes, have proven useful in many scenarios. Software solutions exist that help authors make hypervideos from media files. However, recording and editing video scenes for hypervideos is a tedious and time consuming job. Huge video databases like YouTube exist that can provide rich sources of video material. Yet it is often illegal to download and re-purpose videos from these sites, requiring a solution that links whole videos or parts of videos and plays them in an embedded player. This work presents the SIVA Web Producer, a Chrome extension for the creation of hypervideos consisting of scenes from YouTube videos. After creating a project, the SIVA Web Producer embeds YouTube videos or parts thereof as video clips. These can then be linked in a scene graph and extended with annotations. The plug-in provides a preview space for testing the …
Cited by:	1
Resources:	[PDF] from acm.org

:view less

Explaining topical distances using word embeddings

2016

Nils Witt, Christin Seifert, Michael Granitzer
2016 27th International Workshop on Database and Expert Systems Applications (DEXA), 21…
:read more

Authors:	Nils Witt, Christin Seifert, Michael Granitzer
Publication date:	09/05/2016
Journal:	2016 27th International Workshop on Database and Expert Systems Applications (DEXA)
Pages:	212-217
Publisher:	IEEE
Description:	Word and document embeddings have gained a lot of attention recently, because they tend to work well in text mining tasks. Yet, they elude humans intuition. In this paper we are making the attempt to explain the arithmetic difference between two document embeddings by a series of word embeddings. We present an algorithm that iteratively picks words from a vocabulary that closes the topical gap between the documents. Moreover, we present the Econstor16 corpus that was used for the experiments. Although not all words that are found are great matches, the algorithm is able to find sets of words that are reasonable to a human that reads both documents. Remarkably, some of the well-explaining words are mentioned in neither documents.
Cited by:	6
Resources:	[PDF] from webis.de

:view less

Supporting Web surfers in finding related material in digital library repositories

2016

Jörg Schlötterer, Christin Seifert, Michael Granitzer
International Conference on Theory and Practice of Digital Libraries, 434-437
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Michael Granitzer
Publication date:	08/10/2016
Journal:	International Conference on Theory and Practice of Digital Libraries
Pages:	434-437
Publisher:	Springer International Publishing
Description:	Web surfers often face the need for additional information beyond the page they are currently reading. While such related material is available in digital library repositories, finding it within these repositories can be a challenging task. In order to ease the burden for the user, we present an approach to construct queries automatically from a textual paragraph. Named entities from the paragraph and a query scheme, which includes the topic of the paragraph form the two pillars of this approach, which is applicable to any search system, that supports keyword queries. Evaluation results point towards users not being able to find optimal queries and needing support in doing so.
Cited by:	7
Resources:	[PDF] from github.io

:view less

Robust and collective entity disambiguation through semantic embeddings

2016

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Proceedings of the 39th International ACM SIGIR conference on Research and Development …
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	07/07/2016
Journal:	Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval
Pages:	425-434
Description:	Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Semantic Search, Reasoning and Question & Answering. We propose a new collective, graph-based disambiguation algorithm utilizing semantic entity and document embeddings for robust entity disambiguation. Robust thereby refers to the property of achieving better than state-of-the-art results over a wide range of very different data sets. Our approach is also able to abstain if no appropriate entity can be found for a specific surface form. Our evaluation shows, that our approach achieves significantly (>5%) better results than all other publicly available disambiguation algorithms on 7 of 9 datasets …
Cited by:	124
Resources:	[PDF] from acm.org

:view less

Doser-a knowledge-base-agnostic framework for entity disambiguation using semantic embeddings

2016

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
European semantic web conference, 182-198
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	05/14/2016
Journal:	European semantic web conference
Pages:	182-198
Publisher:	Springer International Publishing
Description:	Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Semantic Search, Reasoning and Question & Answering. In this work, we propose DoSeR (Disambiguation of Semantic Resources), a (named) entity disambiguation framework that is knowledge-base-agnostic in terms of RDF (e.g. DBpedia) and entity-annotated document knowledge bases (e.g. Wikipedia). Initially, our framework automatically generates semantic entity embeddings given one or multiple knowledge bases. In the following, DoSeR accepts documents with a given set of surface forms as input and collectively links them to an entity in a knowledge base with a graph-based approach. We …
Cited by:	73
Resources:	[PDF] from semanticscholar.org

:view less

Extracting Navigation Hierarchies from Networks with Genetic Algorithms

2016

Stefan John, Michael Granitzer, Denis Helic
International Conference on Web Information Systems and Technologies 3, 63-74
:read more

Authors:	Stefan John, Michael Granitzer, Denis Helic
Publication date:	04/23/2016
Journal:	International Conference on Web Information Systems and Technologies
Volume:	3
Pages:	63-74
Publisher:	SCITEPRESS
Description:	Information networks are nowadays an important source of knowledge, indispensable for our daily tasks. Because of their size, however, efficient navigation can be a challenge. Following the idea to use network hierarchies as guidance in human as well as algorithmic search processes, this work focuses on the creation of optimized navigation hierarchies. Based on an established model of human navigation, decentralized search, we defined two quality criteria for network hierarchies and propose a genetic algorithm applying them. We conducted experiments on an information as well as a social network and analyzed the optimization effectivity of our approach. Furthermore, we investigated the structure of the resulting navigation hierarchies. We found our algorithm to be well-suited for the task of hierarchy optimization and found distinct structural properties influencing the quality of navigational hierarchies.
Resources:	[PDF] from scitepress.org

:view less

Inferring semantic interest profiles from Twitter followees: Does Twitter know better than your friends?

2016

Christoph Besel, Jörg Schlötterer, Michael Granitzer
Proceedings of the 31st Annual ACM symposium on applied computing, 1152-1157
:read more

Authors:	Christoph Besel, Jörg Schlötterer, Michael Granitzer
Publication date:	04/04/2016
Journal:	Proceedings of the 31st Annual ACM symposium on applied computing
Pages:	1152-1157
Description:	Social media based recommendation systems infer users' interests from their social network activity in order to provide personalised recommendations. Typically, the user profiles are generated by analysing the users' posts or tweets. However, there might be a significant difference between what a user produces and what she consumes. We propose an approach for inferring user interests from followees (the accounts the user follows) rather than tweets. This is done by extracting named entities from a user's followees using the English Wikipedia as knowledge base and regarding them as interests. Afterwards, a spreading activation algorithm is performed on a Wikipedia category taxonomy to aggregate the various interests to a more abstract interest profile. With over 7 out of 10 items being relevant to the users in our evaluation, we show that this approach can compete with the state of the art and performs even …
Cited by:	23
Resources:	[PDF] from acm.org

:view less

Evaluating memory efficiency and robustness of word embeddings

2016

Johannes Jurgovsky, Michael Granitzer, Christin Seifert
European conference on information retrieval, 200-211
:read more

Authors:	Johannes Jurgovsky, Michael Granitzer, Christin Seifert
Publication date:	03/20/2016
Journal:	European conference on information retrieval
Pages:	200-211
Publisher:	Springer International Publishing
Description:	Skip-Gram word embeddings, estimated from large text corpora, have been shown to improve many NLP tasks through their high-quality features. However, little is known about their robustness against parameter perturbations and about their efficiency in preserving word similarities under memory constraints. In this paper, we investigate three post-processing methods for word embeddings to study their robustness and memory efficiency. We employ a dimensionality-based, a parameter-based and a resolution-based method to obtain parameter-reduced embeddings and we provide a concept that connects the three approaches. We contrast these methods with the relative accuracy loss on six intrinsic evaluation tasks and compare them with regard to the memory efficiency of the reduced embeddings. The evaluation shows that low Bit-resolution embeddings offer great potential for memory savings by alleviating …
Cited by:	18
Resources:	[PDF] from hal.science

:view less

Taking cultural and scientific content to users through the EEXCESS project

2016

Michael Granitzer, Christin Seifert
D-Lib Magazine 22 (3/4)
:read more

Authors:	Michael Granitzer, Christin Seifert
Publication date:	03/01/2016
Journal:	D-Lib Magazine
Volume:	22
Issue:	3/4
Publisher:	Corporation for National Research Initiatives
Description:	Cultural and scientific content is available in different repositories, distributed within archives, libraries and museum collections across Europe. These collections encompass different kinds of data, among them scientific publications, audio files, images and videos. While such cultural and scientific content is usually openly accessible, accessing it takes efforts and requires in-depth knowledge about the existing repositories and portals–a major hurdle that not all users can successfully overcome. Therefore, EEXCESS pursues the idea of giving users effortless access to Europe's rich digitally available, cultural heritage content. Through contextualized and personalized recommendations of cultural and scientific content we aim to bring the content to the user instead of bringing the users to the content.
Cited by:	1

:view less

D1. 3–Third Conceptual Architecture and Requirements Definition

2016

Michael Granitzer, Christin Seifert, Jörg Schlötterer, Thomas Orgel, Thomas Cerques, Ro…
:read more

Authors:	Michael Granitzer, Christin Seifert, Jörg Schlötterer, Thomas Orgel, Thomas Cerques, Roman Kern, Vedran Sabol, Louis Gantner
Publication date:	01/01/2016
Description:	D1.3 -- Third Conceptual Architecture and Requirements Definition - University of Twente Research Information Skip to main navigation Skip to search Skip to main content University of Twente Research Information Home University of Twente Research Information Logo Search content at University of Twente Research Information Home Profiles Research units Projects Research output Datasets Activities Prizes Press/Media D1.3 -- Third Conceptual Architecture and Requirements Definition Michael Granitzer , Christin Seifert , Jörg Schlötterer , Thomas Orgel , Thomas Cerques , Roman Kern , Vedran Sabol , Louis Gantner Research output: Working paper Overview Original language English Publication status Published - 2016 Cite this APA Author BIBTEX Harvard Standard RIS Vancouver Powered by Pure Link opens in a new tab, Scopus Link opens in a new tab & Elsevier Fingerprint Engine™ Link opens in a new …

:view less

From Single Screen to Dual Screen-a Design study for a User-Controlled Hypervideo-Based Physiotherapy Training.

2016

Britta Meixner, Christian Handschigl, Stefan John, Michael Granitzer
WSICC@ TVX
:read more

Authors:	Britta Meixner, Christian Handschigl, Stefan John, Michael Granitzer
Publication date:	01/01/2016
Journal:	WSICC@ TVX
Description:	Hypervideo based physiotherapy trainings bear an opportunity to support patients in continuing their training after being released from a rehabilitation clinic. Many exercises require the patient to sit on the floor or a gymnastic ball, lie on a gymnastics mat, or do the exercises in other postures. Using a laptop or tablet with a stand to show the exercises is more helpful than for example just having some drawings on a leaflet. However, it may lead to incorrect execution of the exercises while maintaining eye contact with the screen or require the user to get up and select the next exercise if the devices is positioned for a better view. A dual screen application, where contents are shown on a TV screen and the flow of the video can be controlled from a mobile second device, allows patients to keep their correct posture and the same time view and select contents. In this paper we propose first studies for user interface designs for such apps. Initial paper prototypes are discussed and refined in two focus groups. The results are then presented to a broader range of users in a survey. Three prototypes for the mobile app and one prototype for the TV are identified for future user tests.
Cited by:	2
Resources:	[PDF] from academia.edu

:view less

Data-transformation on historical data using the RDF data cube vocabulary

2015

Sebastian Bayerl, Michael Granitzer
Proceedings of the 15th International Conference on Knowledge Technologies and Data-dri…
:read more

Authors:	Sebastian Bayerl, Michael Granitzer
Publication date:	10/21/2015
Journal:	Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business
Pages:	1-8
Description:	This work describes a Data-Integration process, in particular, how XML-based TEI documents with statistical data can be normalized, converted and enriched using the RDF Data Cube Vocabulary. Due to the their inherent complex structure, a simple row-by-row based conversion approach is not sufficient. Therefore, the complex structured tables must be validated, modified and transformed, until they are suitable for the standardized multi-dimensional data structure - the data cube. This paper will especially focus on the transformations necessary to normalize the structure of the tables. A second focus is on the advantages, which come by using the RDF Data Cube Vocabulary. Also, a research prototype was implemented to execute the workflow and convert the statistical data into data cubes.
Cited by:	8
Resources:	[PDF] from acm.org

:view less

A platform for contextual multimedia data: towards a unified metadata model and querying

2015

Kai Schlegel, Emanuel Berndl, Michael Granitzer, Harald Kosch, Thomas Kurz
Proceedings of the 15th International Conference on Knowledge Technologies and Data-dri…
:read more

Authors:	Kai Schlegel, Emanuel Berndl, Michael Granitzer, Harald Kosch, Thomas Kurz
Publication date:	10/21/2015
Journal:	Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business
Pages:	1-8
Description:	Whereas the former Web mostly consisted of information represented in textual documents, nowadays the Web includes a huge number of multimedia documents like videos, photos, and audio. This enormous increase in volume in the private, and above all in the industry sector, makes it more and more difficult to find relevant information. Besides the pure management of multimedia documents, finding hidden semantics and interconnections of heterogeneous cross-media content is a crucial task and stays mostly untouched. To overcome this tendency we see the need for a generic cross-media analysis platform, ranging from extracting relevant features from media objects over representing and publishing extraction results to integrated querying of aggregated findings. In this paper we propose the underlying foundation for a common and contextual multimedia platform in terms of an unified model for publishing …
Cited by:	7
Resources:	[PDF] from acm.org

:view less

Search-based entity disambiguation with document-centric knowledge bases

2015

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Proceedings of the 15th International Conference on Knowledge Technologies and Data-dri…
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	10/21/2015
Journal:	Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business
Pages:	1-8
Description:	Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. One possibility to describe these entities within a knowledge base is via entity-annotated documents (document-centric knowledge base). It has been shown that entity disambiguation with search-based algorithms that use document-centric knowledge bases perform well on the biomedical domain. In this context, the question remains how the quantity of annotated entities within documents and the document count used for entity classification influence disambiguation results. Another open question is whether disambiguation results hold true on more general knowledge data sets (e.g. Wikipedia). In our work we implement a search-based, document-centric disambiguation system and explicitly evaluate the mentioned issues on the biomedical data set CALBC and general knowledge data set …
Cited by:	5
Resources:	[PDF] from acm.org

:view less

From general to specialized domain: Analyzing three crucial problems of biomedical entity disambiguation

2015

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
International Conference on Data Management in Cloud, Grid and P2P Systems, 76-93
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	08/11/2015
Journal:	International Conference on Data Management in Cloud, Grid and P2P Systems
Pages:	76-93
Publisher:	Springer International Publishing
Description:	Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. Most disambiguation systems focus on general purpose knowledge bases like DBpedia but leave out the question how those results generalize to more specialized domains. This is very important in the context of Linked Open Data, which forms an enormous resource for disambiguation. We implement a ranking-based (Learning To Rank) disambiguation system and provide a systematic evaluation of biomedical entity disambiguation with respect to three crucial and well-known properties of specialized disambiguation systems. These are (i) entity context, i.e. the way entities are described, (ii) user data, i.e. quantity and quality of externally disambiguated entities, and (iii) quantity and heterogeneity of entities to disambiguate, i.e. the number and size of different domains in a knowledge base …
Cited by:	11
Resources:	[PDF] from github.io

:view less

User Interface Considerations for Browser-Based Just-in-Time-Retrieval

2015

Christin Seifert, Jörg Schlötterer, Michael Granitzer
2015 19th International Conference on Information Visualisation, 460-467
:read more

Authors:	Christin Seifert, Jörg Schlötterer, Michael Granitzer
Publication date:	07/22/2015
Journal:	2015 19th International Conference on Information Visualisation
Pages:	460-467
Publisher:	IEEE
Description:	With the availability of free online enrichment services injection of additional, external resources in existing Web content becomes more and more widespread. For the specific area of just-in-time retrieval of digital resources based on web page content, there are no specific guidelines of how to design and integrate the additional user interface components. In this paper, we conceptualise related user interface issues, investigating the central questions: (i) how can a user be visually notified that additional results are available, and (ii) with which user interface elements should the results be presented. Concretely, we identified four different notification styles and six different result presentation styles. In a survey-based study with 75 participants we elicited the users' preferences, revealing a clear preference for the representation style (split pane) and a strong preference for three notification styles (notification bubble …
Cited by:	1
Resources:	[PDF] from github.io

:view less

Bacon: Linked Data Integration Based on the RDF Data Cube Vocabulary

2015

Sebastian Bayerl, Michael Granitzer
Proceedings of the 5th International Conference on Web Intelligence, Mining and Semanti…
:read more

Authors:	Sebastian Bayerl, Michael Granitzer
Publication date:	07/13/2015
Journal:	Proceedings of the 5th International Conference on Web Intelligence, Mining and Semantics
Pages:	1-6
Description:	Discovering and integrating relevant real-live datasets are essential tasks, when it comes to handling Linked Data. Similar to Data Warehousing approaches, Linked Data can be prepared to enable sophisticated data analysis. The developed open source framework bacon enables interactive and crowed-sourced Data Integration on Linked Data (Linked Data Integration), utilizing the RDF Data Cube Vocabulary and the semantic properties of Linked Open Data. Discovering suitable datasets on-the-fly in local or remote repositories sets up the ensuing integration process. Based on well-known Data Warehousing processes, the semantic nature of the data is taken into account to handle and merge RDF Data Cubes. To do so, structure and content of the cubes must be analyzed and processed. A similarity measure has been developed to find similarly structured cubes. The user is offered a graphical interface, where …
Cited by:	4
Resources:	[PDF] from acm.org

:view less

The effect of non-linear structures on the usage of hypervideo for physical training

2015

Katrin Tonndorf, Christian Handschigl, Julian Windscheid, Harald Kosch, Michael Granitzer
2015 IEEE International Conference on Multimedia and Expo (ICME), 1-6
:read more

Authors:	Katrin Tonndorf, Christian Handschigl, Julian Windscheid, Harald Kosch, Michael Granitzer
Publication date:	06/29/2015
Journal:	2015 IEEE International Conference on Multimedia and Expo (ICME)
Pages:	1-6
Publisher:	IEEE
Description:	The growing number of elderly people combined with financial cuts in the health care sector lead to an increased demand for computer supported medical services. New standards like HTML5 allow the creation of hypervideo training applications that run on a variety of end user devices. In this paper, we evaluate an HTML5 player running an e-health hypervideo for the support of pelvic floor exercises. In an experimental test setting we compared the hypervideo to a primarily linear version regarding usability and utilization for self-controlled training. Our results show the hypervideo version leads to slightly more usability problems but facilitated a more active and individual training.
Cited by:	10
Resources:	[PDF] from researchgate.net

:view less

Comparing recursive autoencoder and convolutional network for phrase-level sentiment polarity classification

2015

Johannes Jurgovsky, Michael Granitzer
International Conference on Applications of Natural Language to Information Systems, 16…
:read more

Authors:	Johannes Jurgovsky, Michael Granitzer
Publication date:	06/04/2015
Journal:	International Conference on Applications of Natural Language to Information Systems
Pages:	160-166
Publisher:	Springer International Publishing
Description:	We present a comparative evaluation of two neural network architectures, which can be used to compute representations of phrases or sentences. The Semi-Supervised Recursive Autoencoder (SRAE) and the Convolutional Neural Network (CNN) are both methods that directly operate on sequences of words represented via word embeddings and jointly model the syntactic and semantic peculiarities of phrases. We compare both models with respect to their classification accuracy on the task of binary sentiment polarity classification. Our evaluation shows that a single-layer CNN produces equally accurate phrase representations and that both methods profit from the initialization with word embeddings trained by a language model. We observe that the initialization with domain specific word embeddings has no significant effect on the accuracy of both phrase models. A pruning experiment revealed that up to 95 …
Cited by:	3
Resources:	[PDF] from researchgate.net

:view less

Towards a feature-rich data set for personalized access to long-tail content

2015

Christin Seifert, Jörg Schlötterer, Michael Granitzer
Proceedings of the 30th Annual ACM Symposium on Applied Computing, 1031-1038
:read more

Authors:	Christin Seifert, Jörg Schlötterer, Michael Granitzer
Publication date:	04/13/2015
Journal:	Proceedings of the 30th Annual ACM Symposium on Applied Computing
Pages:	1031-1038
Description:	Personalized data access has become one of the core challenges for intelligent information access, especially for nonmainstream long-tail content, as can be found in digital libraries. One of the main reasons that personalization remains a difficult task is the lack of standardized test corpora. In this paper we provide a comprehensive analysis of feature requirements for personalization together with a data collection tool for generating user models and collecting data for personalization of search and recommender system optimization in the long-tail. Based on the feature analysis, we provide a feature-rich publicly available data set, covering web content consumption and creation tasks. Our data set contains user models for eight users, including performed tasks, relevant topics for each task, relevance ratings, and relations between focus text and search queries. Altogether, the data set consists of 217 tasks, 4562 …
Cited by:	12
Resources:	[PDF] from acm.org

:view less

EEXCESS: Personalisierter Zugriff auf Long-Tail-Inhalte-neue Methoden zur Verbreitung wissenschaftlich-kulturellen Wissens.

2015

Christin Seifert, Michael Granitzer
Inf. Wiss. Prax. 66 (2-3), 103-110
:read more

Authors:	Christin Seifert, Michael Granitzer
Publication date:	04/01/2015
Journal:	Inf. Wiss. Prax.
Volume:	66
Issue:	2-3
Pages:	103-110
Description:	Das im folgenden Artikel beschriebene EU FP 7 Projekt EEXCESS untersucht neue Möglichkeiten zur Verbreitung wissenschaftlich-‐kultureller Inhalte digitaler Bibliotheken. Im Zentrum stehen dabei intelligente, die Privatsphäre erhaltende, personalisierte Empfehlungstechniken für wissenschaftliche und kulturelle Long-‐Tail-‐Inhalte. Diese Inhalte sollen dabei in häufig genutzte Web-‐Kanäle automatisiert injiziert und somit einfacher nutzbar und für eine breitere Anwenderbasis sichtbar gemacht werden. EEXCESS bringt die Inhalte zu den Nutzern und Nutzerinnen, anstatt Nutzer und Nutzerinnen zu den Inhalten zu führen–so das Ziel.
Cited by:	2
Resources:	[PDF] from researchgate.net

:view less

From context-aware to context-based: Mobile just-in-time retrieval of cultural heritage objects

2015

Jörg Schlötterer, Christin Seifert, Wolfgang Lutz, Michael Granitzer
European Conference on Information Retrieval, 805-808
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Wolfgang Lutz, Michael Granitzer
Publication date:	03/29/2015
Journal:	European Conference on Information Retrieval
Pages:	805-808
Publisher:	Springer International Publishing
Description:	Cultural content providers face the challenge of disseminating their content to the general public. Meanwhile, access to Web resources shifts from desktop to mobile devices and the wide range of contextual sensors of those devices can be used to proactively retrieve and present resources in an unobtrusive manner. This proactive process, also known as just-in-time retrieval, increases the amount of information viewed and hence is a viable way to increase the visibility of cultural content. We provide a contextual model for mobile just-in-time retrieval, discuss the role of sensor information for its contextual dimensions and show the model’s applicability with a prototypical implementation. Our proposed approach enriches a user’s web experience with cultural content and the developed model can provide guidance for other domains.
Cited by:	8
Resources:	[PDF] from github.io

:view less

Linking biomedical data to the cloud

2015

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Smart Health: Open Problems and Future Challenges, 209-235
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	02/25/2015
Journal:	Smart Health: Open Problems and Future Challenges
Pages:	209-235
Publisher:	Springer International Publishing
Description:	The application of Knowledge Discovery and Data Mining approaches forms the basis of realizing the vision of Smart Hospitals. For instance, the automated creation of high-quality knowledge bases from clinical reports is important to facilitate decision making processes for clinical doctors. A subtask of creating such structured knowledge is entity disambiguation that establishes links by identifying the correct semantic meaning from a set of candidate meanings to a text fragment. This paper provides a short, concise overview of entity disambiguation in the biomedical domain, with a focus on annotated corpora (e.g. CalbC), term disambiguation algorithms (e.g. abbreviation disambiguation) as well as gene and protein disambiguation algorithms (e.g. inter-species gene name disambiguation). Finally, we provide some open problems and future challenges that we expect future research will take into account.
Cited by:	5
Resources:	[PDF] from github.io

:view less

D1. 2–Second Conceptual Architecture and Requirements Definition

2015

Michael Granitzer, Christin Seifert, Jörg Schlötterer, Thomas Orgel, Silvia Russegger, …
:read more

Authors:	Michael Granitzer, Christin Seifert, Jörg Schlötterer, Thomas Orgel, Silvia Russegger, Kris Jack, Nils Witt, Timo Borst, Gordon McKenna, Louis Gantner, Gerhard Doppler, Vedran Sabol, Hermann Ziak, Christopher Koska, Thomas Cerques
Publication date:	01/01/2015
Description:	D1.2 -- Second Conceptual Architecture and Requirements Definition - University of Twente Research Information Skip to main navigation Skip to search Skip to main content University of Twente Research Information Home University of Twente Research Information Logo Home Profiles Research units Projects Research output Datasets Activities Prizes Press/Media Search by expertise, name or affiliation D1.2 -- Second Conceptual Architecture and Requirements Definition Michael Granitzer, Christin Seifert, Jörg Schlötterer, Thomas Orgel, Silvia Russegger, Kris Jack, Nils Witt, Timo Borst, Gordon McKenna, Louis Gantner, Gerhard Doppler, Vedran Sabol, Hermann Ziak, Christopher Koska, Thomas Cerques Datamanagement & Biometrics Research output: Working paper Overview Original language English Publisher Passau University Publication status Published - 2015 Cite this APA Author BIBTEX Harvard …

:view less

TIR 2015 Workshop Preface

2015

Michael Granitzer, Christin Seifert, Benno Stein
26th International Conference on Database and Expert Systems Applications, DEXA 2015
:read more

Authors:	Michael Granitzer, Christin Seifert, Benno Stein
Publication date:	01/01/2015
Journal:	26th International Conference on Database and Expert Systems Applications, DEXA 2015
Publisher:	IEEE
Description:	Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

:view less

Digital library content in the social web: Resource usage and content injection

2015

Christin Seifert, Nils Witt, Sebastian Bayerl, Michael Granitzer
STCSN Newsletter 3 (1)
:read more

Authors:	Christin Seifert, Nils Witt, Sebastian Bayerl, Michael Granitzer
Publication date:	01/01/2015
Journal:	STCSN Newsletter
Volume:	3
Issue:	1
Publisher:	IEEE
Description:	Providers of digital non-mainstream content face two major challenges. First, they have little knowledge about the usage of their content in social media channels, which is neccesary to improve dissemination strategies. The second challenge is how to bring the content to interested users, who do not know their portals. We present two case studies for analysing and injecting content into social media channels. More specifically, we analyse the usage of scientific literature from the domain of economics in blogs and tweets that are related to the field of economics. Additionally, we present two mechanisms for injecting content into blogs or tweets, by the means of a Wordpress plugin and a Twitter bot respectively. The usage analysis shows that the resource coverage is rather low in the investigated social media channels (≈ 0.15% in tweets and≈ 0.5% in blogs). Using a Twitter bot for content dissemination is feasible, but not advisable because the automatic approaches are quite hard to optimise towards Twitter’s usage policy. For content injection into the blogosphere the Wordpress plugin shows promising results in a qualitative user study. Our results indicate that content injection into social media channels is a promising way for content dissemination and point towards a preference of blogs over twitter as target channel.
Cited by:	4
Resources:	[PDF] from utwente.nl

:view less

A game with a purpose to access Europe’s cultural treasure

2015

Jörg Schlötterer, Christin Seifert, Lisa Wagner, Michael Granitzer
2nd International Workshop on Gamification for Information Retrieval, GamifIR 2015, 13-18
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Lisa Wagner, Michael Granitzer
Publication date:	01/01/2015
Journal:	2nd International Workshop on Gamification for Information Retrieval, GamifIR 2015
Pages:	13-18
Publisher:	CEUR
Description:	Europeana, the European aggregator for digital cultural heritage objects spends huge effort on providing access to the vast collection of Europe’s Culture. While the content has been made available online, it still needs to be disseminated. We propose a “games with a purpose” approach to engage people with this tremendous collection and make them discover European culture. This approach is implemented by a question and answering system, in which players create questions that are answered by specific Europeana resources. Other players can then use a search interface to Europeana to find the particular resource that is needed to answer the question. This concept provides a low level entrance to cultural heritage for end users. Moreover it reveals human search strategies on Europeana that can be exploited to support and improve the search experience of other users on Europeana and helps to identify objects of interest by analyzing usage data.
Cited by:	9
Resources:	[PDF] from utwente.nl

:view less

Collaborative development of a semantic wiki on forest management decision support

2014

AF Marques, C Rosset, Jussi Rasinmaki, H Vacik, S Gordon, S Nobre, A Falcao, D Weber, M…
Scandinavian Journal of Forest Research 29 (sup1), 30-43
:read more

Authors:	AF Marques, C Rosset, Jussi Rasinmaki, H Vacik, S Gordon, S Nobre, A Falcao, D Weber, M Granitzer, LO Eriksson
Publication date:	11/01/2014
Journal:	Scandinavian Journal of Forest Research
Volume:	29
Issue:	sup1
Pages:	30-43
Publisher:	Taylor & Francis
Description:	Semantic wikis support collaboratively editing, categorising, interlinking and retrieving web pages for a group of experts working in a certain domain. The use of semantic technologies allows the expression of wiki content in a more structured way, which increases its potential use. This contribution presents an overview of the development process towards a semantic wiki related to a repository of forest decision support systems, including models, methods and data used, as well as case studies and lessons learned. An international group of experts took part in the conceptualisation of the semantic wiki (i.e. identification of wiki properties and forms), provided content and developed queries to analyse the information gathered. The resulting ForestDSS wiki gives an overview of the current use, development and application of forest decision support systems worldwide. Based on the experiences gathered during the …
Cited by:	14

:view less

Linked data warehousing

2014

Sebastian Bayerl, Michael Granitzer
Linked Enterprise Data: Management und Bewirtschaftung vernetzter Unternehmensdaten mit…
:read more

Authors:	Sebastian Bayerl, Michael Granitzer
Publication date:	10/31/2014
Journal:	Linked Enterprise Data: Management und Bewirtschaftung vernetzter Unternehmensdaten mit Semantic Web Technologien
Pages:	177-192
Publisher:	Springer Berlin Heidelberg
Description:	Data-Warehousing bezeichnet die technologische Realisierung analytischer Datenbestände sowie entsprechender Schnittstellen zu deren Exploration und Analyse. Linked Data bietet vor allem mit der vor Kurzem begonnenen Entwicklung des RDF Data Cube Vokabulars neue Entwicklungsmöglichkeiten für Data-Warehousing Technologien und deren Einsatzspektrum. Der Beitrag stellt die Grundlagen zu Data-Warehouses vor und führt in das RDF Data Cube Vokabular als Linked Data Äquivalent ein. Beide Grundlagen dienen der Diskussion sowohl der Anwendung von RDF Data Cubes im Data-Warehousing als auch der Erweiterung traditioneller Data-Warehousing Ansätze, z. B. durch Integration offener Daten in Data-Warehousing Prozessen.
Cited by:	3

:view less

Balloon Synopsis: A jQuery plugin to easily integrate the Semantic Web in a website.

2014

Kai Schlegel, Thomas Weißgerber, Florian Stegmaier, Michael Granitzer, Harald Kosch
ISWC (Developers Workshop), 19-24
:read more

Authors:	Kai Schlegel, Thomas Weißgerber, Florian Stegmaier, Michael Granitzer, Harald Kosch
Publication date:	10/19/2014
Journal:	ISWC (Developers Workshop)
Pages:	19-24
Description:	The Semantic Web grows constantly and promises a huge amount of machine-interpretable information. Unfortunately, the integration and usage of semantic information is not feasible by everyone. Hence, a large number of Semantic Web applications are lacking and the potential of the semantic knowledge remains unexploited. We propose balloon Synopsis, an easy-to-use jQuery plugin to integrate Semantic Web information in a website. It provides a modern visualisation and browser for RDF information including automatic remote information enhancing, similarity analysis and ontology templates. Balloon Synopsis enables web developers to rely on known tools and programming language to benefit of the global knowledge graph.
Cited by:	10
Resources:	[PDF] from ceur-ws.org

:view less

Discovery and visual analysis of linked data for humans

2014

Vedran Sabol, Gerwald Tschinkel, Eduardo Veas, Patrick Hoefler, Belgin Mutlu, Michael G…
International Semantic Web Conference, 309-324
:read more

Authors:	Vedran Sabol, Gerwald Tschinkel, Eduardo Veas, Patrick Hoefler, Belgin Mutlu, Michael Granitzer
Publication date:	10/19/2014
Journal:	International Semantic Web Conference
Pages:	309-324
Publisher:	Springer International Publishing
Description:	Linked Data has grown to become one of the largest available knowledge bases. Unfortunately, this wealth of data remains inaccessible to those without in-depth knowledge of semantic technologies. We describe a toolchain enabling users without semantic technology background to explore and visually analyse Linked Data. We demonstrate its applicability in scenarios involving data from the Linked Open Data Cloud, and research data extracted from scientific publications. Our focus is on the Web-based front-end consisting of querying and visualisation tools. The performed usability evaluations unveil mainly positive results confirming that the Query Wizard simplifies searching, refining and transforming Linked Data and, in particular, that people using the Visualisation Wizard quickly learn to perform interactive analysis tasks on the resulting Linked Data sets. In making Linked Data analysis effectively accessible …
Cited by:	38
Resources:	[PDF] from githubusercontent.com

:view less

Balloon Synopsis: A Modern Node-Centric RDF Viewer and Browser for the Web

2014

Michael Granitzer, Harald Kosch
The Semantic Web: ESWC 2014 Satellite Events: ESWC 2014 Satellite Events, Anissaras, Cr…
:read more

Authors:	Michael Granitzer, Harald Kosch
Publication date:	10/15/2014
Journal:	The Semantic Web: ESWC 2014 Satellite Events: ESWC 2014 Satellite Events, Anissaras, Crete, Greece, May 25-29, 2014, Revised Selected Papers
Pages:	249
Publisher:	Springer
Description:	Nowadays, the RDF data model is a crucial part of the Semantic Web. Especially web developers favour RDF serialization formats like RDFa and JSON-LD. However, the visualization of large portions of RDF data in an appealing way is still a cumbersome task. RDF visualizers in general are not targeting the Web as usage scenario or simply display the complex RDF graph directly rather than applying a human friendly facade. Balloon Synopsis tries to overcome these issues by providing an easy-to-use RDF visualizer based on HTML and JavaScript. For an ease integration, it is implemented as jQuery-plugin offering a node-centric RDF viewer and browser with automatic Linked Data enhancement in a modern tile design.

:view less

A multimedia help system for a medical scenario in a rehabilitation clinic

2014

Britta Meixner, Katrin Tonndorf, Stefan John, Christian Handschigl, Kai Hofmann, Michae…
Proceedings of the 14th International Conference on Knowledge Technologies and Data-dri…
:read more

Authors:	Britta Meixner, Katrin Tonndorf, Stefan John, Christian Handschigl, Kai Hofmann, Michael Granitzer
Publication date:	09/16/2014
Journal:	Proceedings of the 14th International Conference on Knowledge Technologies and Data-driven Business
Pages:	1-8
Description:	In this paper we outline instructional, legal, and software requirements as well as a prototypical software implementation for a multimedia help system in a rehabilitation scenario. The help system will be used by patients in a rehabilitation clinic to support their pelvic floor exercises. After describing the use case we will outline the requirements for the scenario and show how our software was extended to meet the requirements. We present the key features of each component of the SIVA Suite: the authoring tool, the redesigned HTML5 player, and the server application. Thereby, important new features are described and illustrated. Design decisions are presented and explained.
Cited by:	12
Resources:	[PDF] from acm.org

:view less

Recommending scientific literature: Comparing use-cases and algorithms

2014

Roman Kern, Kris Jack, Michael Granitzer
arXiv preprint arXiv:1409.1357
:read more

Authors:	Roman Kern, Kris Jack, Michael Granitzer
Publication date:	09/04/2014
Journal:	arXiv preprint arXiv:1409.1357
Description:	An important aspect of a researcher's activities is to find relevant and related publications. The task of a recommender system for scientific publications is to provide a list of papers that match these criteria. Based on the collection of publications managed by Mendeley, four data sets have been assembled that reflect different aspects of relatedness. Each of these relatedness scenarios reflect a user's search strategy. These scenarios are public groups, venues, author publications and user libraries. The first three of these data sets are being made publicly available for other researchers to compare algorithms against. Three recommender systems have been implemented: a collaborative filtering system; a content-based filtering system; and a hybrid of these two systems. Results from testing demonstrate that collaborative filtering slightly outperforms the content-based approach, but fails in some scenarios. The hybrid system, that combines the two recommendation methods, provides the best performance, achieving a precision of up to 70%. This suggests that both techniques contribute complementary information in the context of recommending scientific literature and different approaches suite for different information needs.
Cited by:	9
Resources:	[PDF] from arxiv.org

:view less

From Tail to Head: Browser Based Suggestion of Long-tail Resources

2014

Jörg Schlötterer, Christin Seifert, Roman Kern, Michael Granitzer
14th International Conference on Knowledge Technologies and Data-driven Business, i-KNO…
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Roman Kern, Michael Granitzer
Publication date:	09/01/2014
Journal:	14th International Conference on Knowledge Technologies and Data-driven Business, i-KNOW 2014
Description:	Cultural, scientific and educational resources often remain buried in deep web portals like libraries or museum archives in the so called long-tail of the web. The reason for these long-tail contents largely remaining untapped is twofold: One factor is that most users are not aware of the existence of these contents, as they do not appear high-ranked on general purpose search engines, due to their lower popularity. The second problem is that the contents are spread widely over the Web, residing in a multitude of small repositories, which cover specific domains.

:view less

Unsupervised document structure analysis of digital scientific articles

2014

Stefan Klampfl, Michael Granitzer, Kris Jack, Roman Kern
International journal on digital libraries 14 (3), 83-99
:read more

Authors:	Stefan Klampfl, Michael Granitzer, Kris Jack, Roman Kern
Publication date:	08/01/2014
Journal:	International journal on digital libraries
Volume:	14
Issue:	3
Pages:	83-99
Publisher:	Springer Berlin Heidelberg
Description:	Text mining and information retrieval in large collections of scientific literature require automated processing systems that analyse the documents’ content. However, the layout of scientific articles is highly varying across publishers, and common digital document formats are optimised for presentation, but lack structural information. To overcome these challenges, we have developed a processing pipeline that analyses the structure a PDF document using a number of unsupervised machine learning techniques and heuristics. Apart from the meta-data extraction, which we reused from previous work, our system uses only information available from the current document and does not require any pre-trained model. First, contiguous text blocks are extracted from the raw character stream. Next, we determine geometrical relations between these blocks, which, together with geometrical and font information, are then used …
Cited by:	49

:view less

FacetScape: A visualization for exploring the search space

2014

Christin Seifert, Johannes Jurgovsky, Michael Granitzer
2014 18th International Conference on Information Visualisation, 94-101
:read more

Authors:	Christin Seifert, Johannes Jurgovsky, Michael Granitzer
Publication date:	07/16/2014
Journal:	2014 18th International Conference on Information Visualisation
Pages:	94-101
Publisher:	IEEE
Description:	Despite advancing search technologies, information overload has not yet been solved. Getting an overview of information or explorative access to information becomes increasingly difficult with the exponentially increasing amount of information. Search result visualizations, especially for faceted browsing, aim at supporting users to find their way through large document collections. We propose Facets cape, a novel visualization for navigation and refinement of search results allowing users to visually construct complex boolean search queries for narrowing down the search space. This visualization combines Voronoi subdivision and a tag cloud representation of the search facets. Further it includes a preview of action (query preview) and interactions to allow users to focus on important aspects of the data for the task at hand. In a comparative user study with 15 users we compared the visualization to a standard …
Cited by:	31
Resources:	[PDF] from github.io

:view less

Assessing the quality of Web content

2014

Elisabeth Lex, Inayat Khan, Horst Bischof, Michael Granitzer
arXiv preprint arXiv:1406.3188
:read more

Authors:	Elisabeth Lex, Inayat Khan, Horst Bischof, Michael Granitzer
Publication date:	06/12/2014
Journal:	arXiv preprint arXiv:1406.3188
Description:	This paper describes our approach towards the ECML/PKDD Discovery Challenge 2010. The challenge consists of three tasks: (1) a Web genre and facet classification task for English hosts, (2) an English quality task, and (3) a multilingual quality task (German and French). In our approach, we create an ensemble of three classifiers to predict unseen Web hosts whereas each classifier is trained on a different feature set. Our final NDCG on the whole test set is 0:575 for Task 1, 0:852 for Task 2, and 0:81 (French) and 0:77 (German) for Task 3, which ranks second place in the ECML/PKDD Discovery Challenge 2010.
Cited by:	7
Resources:	[PDF] from arxiv.org

:view less

Balloon synopsis: A modern node-centric RDF viewer and browser for the web

2014

Kai Schlegel, Thomas Weißgerber, Florian Stegmaier, Christin Seifert, Michael Granitzer…
European Semantic Web Conference, 249-253
:read more

Authors:	Kai Schlegel, Thomas Weißgerber, Florian Stegmaier, Christin Seifert, Michael Granitzer, Harald Kosch
Publication date:	05/25/2014
Journal:	European Semantic Web Conference
Pages:	249-253
Publisher:	Springer International Publishing
Description:	Nowadays, the RDF data model is a crucial part of the Semantic Web. Especially web developers favour RDF serialization formats like RDFa and JSON-LD. However, the visualization of large portions of RDF data in an appealing way is still a cumbersome task. RDF visualizers in general are not targeting the Web as usage scenario or simply display the complex RDF graph directly rather than applying a human friendly facade. Balloon Synopsis tries to overcome these issues by providing an easy-to-use RDF visualizer based on HTML and JavaScript. For an ease integration, it is implemented as jQuery-plugin offering a node-centric RDF viewer and browser with automatic Linked Data enhancement in a modern tile design.
Cited by:	19
Resources:	[PDF] from academia.edu

:view less

D 7.5-Web-enabled Public Showcase

2014

Commercially Empowered
:read more

Authors:	Commercially Empowered
Publication date:	05/02/2014
Description:	This deliverable describes Web-enabled public showcases for public dissemination and presentation of the project. The showcase is built around publicly available, Web-based CODE prototypes. The purpose of each prototype is briefly described followed by typical use case descriptions demonstrated along a typical use case.

:view less

5.3 Monitored Success Factors

2014

Maya Hristakeva, Michael Granitzer, Roman Kern, Michael Hollauf
:read more

Authors:	Maya Hristakeva, Michael Granitzer, Roman Kern, Michael Hollauf
Publication date:	04/30/2014
Description:	This deliverable reports on the success factors monitored by the three prototypes of the CODE platform:

:view less

Balloon fusion: SPARQL rewriting based on unified co-reference information

2014

Kai Schlegel, Florian Stegmaier, Sebastian Bayerl, Michael Granitzer, Harald Kosch
2014 IEEE 30th International Conference on Data Engineering Workshops, 254-259
:read more

Authors:	Kai Schlegel, Florian Stegmaier, Sebastian Bayerl, Michael Granitzer, Harald Kosch
Publication date:	03/31/2014
Journal:	2014 IEEE 30th International Conference on Data Engineering Workshops
Pages:	254-259
Publisher:	IEEE
Description:	While Linked Open Data showed enormous increase in volume, yet there is no single point of access for querying the over 200 SPARQL repositories. In this paper we present Balloon Fusion, a SPARQL 1.1 rewriting and query federation service build on crawling and consolidating co-reference relationships in over 100 reachable Linked Data SPARQL Endpoints. The results of this process are 17.6M co-reference statements that have been clustered to 8.4M distinct semantic entities and are now accessible as download for further analysis. The proposed SPARQL rewriting performs a substitution of all URI occurrences with their synonyms combined with an automatic endpoint selection based on URI origin for a comprehensive query federation. While we show the technical feasibility, we also critically reflect the current status of the Linked Open Data cloud: although it is huge in size, access via SPARQL Endpoints is …
Cited by:	25
Resources:	[PDF] from researchgate.net

:view less

5.2 CODE Platform Beta Version

2014

WP Lead Partner Mendeley, Florian Stegmaier, Michael Granitzer, Patrick Höfler, Roman K…
:read more

Authors:	WP Lead Partner Mendeley, Florian Stegmaier, Michael Granitzer, Patrick Höfler, Roman Kern, Kris Jack, Michael Hollauf, Vedran Sabol
Publication date:	02/05/2014
Description:	The first platform is the Mendeley Desktop Client and the corresponding Mendeley API. The Mendeley Desktop Client uses CODE services that enrich scientific research papers by extracting table of contents, tables, figures and entities for nearly one million open access publications available in Mendeley. The extracted elements are partially available as JSON-LD and freely available via the Mendeley REST API. The features are available to all Mendeley Users and it is the first time that enriched PDFs are available to the scientific community. The second platform is the MindMeister Collaborative mind-mapping platform. Through CODE MindMeister now provides a semi-automatic extension of MindMaps with different Linked Open Data Sources like Freebase. Mind-maps, which constitute thesauri, are exported using the SKOS Vocabular and hence are available for further re-use in the semantic web. For research papers, MindMeister offers a presentation mode on top of enriched research papers and CODEs linked data search and visualisation facilities. The third platform, which we named 42-Data, integrates all CODE services in order to bring Linked Open Data to Linked Data laymen. 42-Data constitutes a data bookmarking and data-centric question and answering portal. Researchers are enabled to explore the Linked Open Data cloud and to collect data from different sources like research papers, or non-semantic open data portals. With the collected data they can engage in discussions using data to underpin their opinion. The portal emphasizes a donation based revenue model. The model has been chosen to allow micropayments among …

:view less

Web-based just-in-time retrieval for cultural content

2014

Jörg Schlötterer, Christin Seifert, Michael Granitzer
7th International Workshop on Personalized Access to Cultural Heritage 2014
:read more

Authors:	Jörg Schlötterer, Christin Seifert, Michael Granitzer
Publication date:	02/01/2014
Journal:	7th International Workshop on Personalized Access to Cultural Heritage 2014
Description:	Digital content providers of cultural resources face the challenge of disseminating their content to interested users–either because users do not know the existence of resources or do not know the access points. We propose to apply just-intime retrieval mechanisms to bring the content to the users in a web-based scenario. Our first prototype is a Chrome extension which proposes Europeana content based on the current user context, ie, the current web site. We think that our approach is promising for scenarios where users are not aware of available content and therefore can and do not explicitly state an information need.
Cited by:	15
Resources:	[PDF] from utwente.nl

:view less

Suggesting visualisations for published data

2014

Belgin Mutlu, Patrick Hoefler, Gerwald Tschinkel, Eduardo Veas, Vedran Sabol, Florian S…
2014 International Conference on Information Visualization Theory and Applications (IVA…
:read more

Authors:	Belgin Mutlu, Patrick Hoefler, Gerwald Tschinkel, Eduardo Veas, Vedran Sabol, Florian Stegmaier, Michael Granitzer
Publication date:	01/05/2014
Journal:	2014 International Conference on Information Visualization Theory and Applications (IVAPP)
Pages:	267-275
Publisher:	IEEE
Description:	Research papers are published in various digital libraries, which deploy their own meta-models and technologies to manage, query, and analyze scientific facts therein. Commonly they only consider the meta-data provided with each article, but not the contents. Hence, reaching into the contents of publications is inherently a tedious task. On top of that, scientific data within publications are hardcoded in a fixed format (e.g. tables). So, even if one manages to get a glimpse of the data published in digital libraries, it is close to impossible to carry out any analysis on them other than what was intended by the authors. More effective querying and analysis methods are required to better understand scientific facts. In this paper, we present the web-based CODE Visualisation Wizard, which provides visual analysis of scientific facts with emphasis on automating the visualisation process, and present an experiment of its …
Cited by:	23
Resources:	[PDF] from scitepress.org

:view less

Suggesting Visualisations for Published Data

2014

Michael Granitzer, Vedran Sabol, Eduardo Veas, Florian Stegmaier, Patrick Hoefler
Proceedings of the 5th International Conference on Information Visualization Theory and…
:read more

Authors:	Michael Granitzer, Vedran Sabol, Eduardo Veas, Florian Stegmaier, Patrick Hoefler
Publication date:	01/01/2014
Journal:	Proceedings of the 5th International Conference on Information Visualization Theory and Applications
Description:	Research papers are published in various digital libraries, which deploy their own meta-models and technologies to manage, query, and analyze scientific facts therein. Commonly they only consider the meta-data provided with each article, but not the contents. Hence, reaching into the contents of publications is inherently a tedious task. On top of that, scientific data within publications are hardcoded in a fixed format (eg tables). So, even if one manages to get a glimpse of the data published in digital libraries, it is close to impossible to carry out any analysis on them other than what was intended by the authors. More effective querying and analysis methods are required to better understand scientific facts. In this paper, we present the web-based CODE Visualisation Wizard, which provides visual analysis of scientific facts with emphasis on automating the visualisation process, and present an experiment of its application. We also present the entire analytical process and the corresponding tool chain, including components for extraction of scientific data from publications, an easy to use user interface for querying RDF knowledge bases, and a tool for semantic annotation of scientific data sets.
Resources:	[HTML] from academia.edu

:view less

First Security Proxy Prototype and Reputation Protocols

2014

Sonia Ben Mokhtar, Nadia Bennani, Lionel Brunie, Thomas Cerqueus, Elöd Egyed-Zsigmond, …
:read more

Authors:	Sonia Ben Mokhtar, Nadia Bennani, Lionel Brunie, Thomas Cerqueus, Elöd Egyed-Zsigmond, Omar Hasan, Albin Petit, Pierre-Edouard Portier, Michael Granitzer, Christin Seifert, Jörg Schlötterer
Publication date:	01/01/2014
Publisher:	EEXCESS Consortium
Description:	One of the main objectives of the EEXCESS project is to recommend relevant resources to users while they are browsing the web. Among other applications, a Google Chrome extension has been developed to achieve this objective. It allows users to get recommendations based on the page they are viewing, the text they are highlighting, or the query they issue. Ideally the system must provide personalised recommendations to satisfy each user’s needs. In order to do that, the system has to collect information about users: user profiles(eg, name, gender or birth date), browsing history, etc. This situation may lead to privacy violations, and thus restraint users to use the system. Work Package 6 works on the privacy aspects of the project. As the deliverable is a prototype, this document is predominantly technical. It reports on the work done on the EEXCESS system and on the Google Chrome extension. The main contributions described in this documents are: • First implementation of the privacy proxy. The privacy proxy will be used to filter the information transmitted from the user to the federated recommender according to the user privacy policy. Its role is to avoid privacy breaches. At this stage, the privacy proxy provides with three features:(i) it forwards queries to the federated recommender(no filtering is required, as it is already done on the client side),(ii) it logs the users’ activity and(iii) it suggests categories to represent users’ topics of interest.
Cited by:	2

:view less

Lost in Semantics? Ballooning the Web of Data

2014

Florian Stegmaier, Kai Schlegel, Michael Granitzer
ERCIM News 2014 (96)
:read more

Authors:	Florian Stegmaier, Kai Schlegel, Michael Granitzer
Publication date:	01/01/2014
Journal:	ERCIM News
Volume:	2014
Issue:	96
Description:	Available from: Florian Stegmaier Retrieved on: 10 July 2015 words, observation parameters and measuring units (bottom part of the picture), it is then possible to populate widgets with content that is tailored to the user's profile (eg, datasets and services applicable to the user’s research area). Also, recourse to a gazetteer for expressing geographic locations simplifies discovery by reducing the need for a map.
Resources:	[PDF] from github.io

:view less

Linked data query wizard: A novel interface for accessing sparql endpoints

2014

Patrick Hoefler, Michael Granitzer, Eduardo Veas, Christin Seifert
7th Workshop on Linked Data on the Web 2014
:read more

Authors:	Patrick Hoefler, Michael Granitzer, Eduardo Veas, Christin Seifert
Publication date:	01/01/2014
Journal:	7th Workshop on Linked Data on the Web 2014
Description:	In an interconnected world, Linked Data is more important than ever before. However, it is still quite difficult to access this new wealth of semantic data directly without having in-depth knowledge about SPARQL and related semantic technologies. Also, most people are currently used to con-suming data as 2-dimensional tables. Linked Data is by defi-nition always a graph, and not that many people are used to handle data in graph structures. Therefore we present the Linked Data Query Wizard, a web-based tool for displaying, accessing, filtering, exploring, and navigating Linked Data stored in SPARQL endpoints. The main innovation of the interface is that it turns the graph structure of Linked Data into a tabular interface and provides easy-to-use interaction possibilities by using metaphors and techniques from current search engines and spreadsheet applications that regular web users are already familiar with.
Cited by:	47
Resources:	[PDF] from utwente.nl

:view less

Visual analysis and knowledge discovery for text

2013

Christin Seifert, Vedran Sabol, Wolfgang Kienreich, Elisabeth Lex, Michael Granitzer
Large-scale data analytics, 189-218
:read more

Authors:	Christin Seifert, Vedran Sabol, Wolfgang Kienreich, Elisabeth Lex, Michael Granitzer
Publication date:	11/28/2013
Journal:	Large-scale data analytics
Pages:	189-218
Publisher:	Springer New York
Description:	Providing means for effectively accessing and exploring large textual data sets is a problem attracting the attention of text mining and information visualization experts alike. The rapid growth of the data volume and heterogeneity, as well as the richness of metadata and the dynamic nature of text repositories, add to the complexity of the task. This chapter provides an overview of data visualization methods for gaining insight into large, heterogeneous, dynamic textual data sets. We argue that visual analysis, in combination with automatic knowledge discovery methods, provides several advantages. Besides introducing human knowledge and visual pattern recognition into the analytical process, it provides the possibility to improve the performance of automatic methods through user feedback.
Cited by:	19
Resources:	[PDF] from researchgate.net

:view less

Proceedings of the 13th International Conference on Knowledge Management and Knowledge Technologies

2013

Stefanie Lindstaedt, Michael Granitzer
:read more

Authors:	Stefanie Lindstaedt, Michael Granitzer
Publication date:	09/04/2013
Publisher:	ACM
Description:	The i-KNOW 2013, organized by Know-Center GmbH and Graz University of Technology is the 13th event in the successful tradition of the conference series on knowledge management and knowledge technologies. Since its inception in 2001 the conference has grown into an internationally highly acknowledged event attracting more than 500 attendees every year.
Cited by:	1

:view less

Do we need entity-centric knowledge bases for entity disambiguation?

2013

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Proceedings of the 13th International Conference on Knowledge Management and Knowledge …
:read more

Authors:	Stefan Zwicklbauer, Christin Seifert, Michael Granitzer
Publication date:	09/04/2013
Journal:	Proceedings of the 13th International Conference on Knowledge Management and Knowledge Technologies
Pages:	1-8
Description:	Entity Disambiguation has been studied extensively in the last 10 years with authors reporting increasingly well performing systems. However, most studies focused on general purpose knowledge bases like Wikipedia or DBPedia and left out the question how those results generalize to more specialized domains. This is especially important in the context of Linked Open Data which forms an enormous resource for disambiguation. However, the influence of domain heterogeneity, size and quality of the knowledge base remains largely unanswered. In this paper we present an extensive set of experiments on special purpose knowledge bases from the biomedical domain where we evaluate the disambiguation performance along four variables: (i) the representation of the knowledge base as being either entity-centric or document-centric, (ii) the size of the knowledge base in terms of entities covered, (iii) the semantic …
Cited by:	23
Resources:	[PDF] from acm.org

:view less

Automated Visualization Support for Linked Research Data.

2013

Belgin Mutlu, Patrick Hoefler, Vedran Sabol, Gerwald Tschinkel, Michael Granitzer
I-SEMANTICS (Posters & Demos) 1026, 40-44
:read more

Authors:	Belgin Mutlu, Patrick Hoefler, Vedran Sabol, Gerwald Tschinkel, Michael Granitzer
Publication date:	08/01/2013
Journal:	I-SEMANTICS (Posters & Demos)
Volume:	1026
Pages:	40-44
Description:	Finding, organizing and analyzing research data (ie publications) published in various digital libraries are often tedious tasks. Each digital library deploys their own meta-model and technology to query and analyze the knowledge (in further text, scientific facts) contained in research publications. The goal of the EU-funded research project CODE is to provide methods for federated querying and analysis of such data. To achieve this, the CODE project offers a platform, that extracts scientific facts from research data and integrates them within the Linked Data Cloud using a common vocabulary (ie meta-model). To support users in analyzing scientific facts, the project provides means for easy-to-use visual analysis. In this paper, we present the web-based CODE Visualization Wizard, which aims to analyze research data visually with an emphasis on automating the visualization process. The main focus of the paper lies on a mapping strategy, which integrates various vocabularies to facilitate the automated visualization process.
Cited by:	19
Resources:	[PDF] from ceur-ws.org

:view less

Crowdsourcing fact extraction from scientific literature

2013

Christin Seifert, Michael Granitzer, Patrick Höfler, Belgin Mutlu, Vedran Sabol, Kai Sc…
International Workshop on Human-Computer Interaction and Knowledge Discovery in Complex…
:read more

Authors:	Christin Seifert, Michael Granitzer, Patrick Höfler, Belgin Mutlu, Vedran Sabol, Kai Schlegel, Sebastian Bayerl, Florian Stegmaier, Stefan Zwicklbauer, Roman Kern
Publication date:	07/01/2013
Journal:	International Workshop on Human-Computer Interaction and Knowledge Discovery in Complex, Unstructured, Big Data
Pages:	160-172
Publisher:	Springer Berlin Heidelberg
Description:	Scientific publications constitute an extremely valuable body of knowledge and can be seen as the roots of our civilisation. However, with the exponential growth of written publications, comparing facts and findings between different research groups and communities becomes nearly impossible. In this paper, we present a conceptual approach and a first implementation for creating an open knowledge base of scientific knowledge mined from research publications. This requires to extract facts - mostly empirical observations - from unstructured texts (mainly PDF’s). Due to the importance of extracting facts with high-accuracy and the impreciseness of automatic methods, human quality control is of utmost importance. In order to establish such quality control mechanisms, we rely on intelligent visual interfaces and on establishing a toolset for crowdsourcing fact extraction, text mining and data integration tasks.
Cited by:	19
Resources:	[HTML] from academia.edu

:view less

Unfolding cultural, educational and scientific long-tail content in the web

2013

Michael Granitzer, Christin Seifert, Silvia Russegger, Klaus Tochtermann
21st International Conference on User Modeling, Adaptation and Personalization, UMAP 2013
:read more

Authors:	Michael Granitzer, Christin Seifert, Silvia Russegger, Klaus Tochtermann
Publication date:	06/01/2013
Journal:	21st International Conference on User Modeling, Adaptation and Personalization, UMAP 2013
Publisher:	CEUR
Description:	This project poster introduces the recently started EEXCESS project, which aims at Enhancing Europes eXchange in Cultural Educational and Scientific Resource. Europe has digitised vast amounts of cultural, scientific and educational content like for example scientific research, historical sound recordings, images of sculptures, films and sheet music. However, since content dissemination on the Web is driven by a small number of large central hubs like social networks or search engines, this cultural and scientific treasures has hardly been recognized by the general public or utilized in scientific and educational processes. EEXCESS aims to develop personalized and contextualized recommendation technologies to augment existing content dissemination channels (eg social media) and content creation process (eg blogging) for distributing high-quality educational, scientific and cultural content. In this project poster we present the underlying idea and related work with focus on user modeling and personalized, context-aware recommendation.
Cited by:	8
Resources:	[PDF] from utwente.nl

:view less

Linked data query wizard: A tabular interface for the semantic web

2013

Patrick Hoefler, Michael Granitzer, Vedran Sabol, Stefanie Lindstaedt
Extended Semantic Web Conference, 173-177
:read more

Authors:	Patrick Hoefler, Michael Granitzer, Vedran Sabol, Stefanie Lindstaedt
Publication date:	05/26/2013
Journal:	Extended Semantic Web Conference
Pages:	173-177
Publisher:	Springer Berlin Heidelberg
Description:	Linked Data has become an essential part of the Semantic Web. A lot of Linked Data is already available in the Linked Open Data cloud, which keeps growing due to an influx of new data from research and open government activities. However, it is still quite difficult to access this wealth of semantically enriched data directly without having in-depth knowledge about SPARQL and related semantic technologies. In this paper, we present the Linked Data Query Wizard, a prototype that provides a Linked Data interface for non-expert users, focusing on keyword search as an entry point and a tabular interface providing simple functionality for filtering and exploration.
Cited by:	17

:view less

Trusted facts: Triplifying primary research data enriched with provenance information

2013

Kai Schlegel, Sebastian Bayerl, Stefan Zwicklbauer, Florian Stegmaier, Christin Seifert…
Extended Semantic Web Conference, 268-270
:read more

Authors:	Kai Schlegel, Sebastian Bayerl, Stefan Zwicklbauer, Florian Stegmaier, Christin Seifert, Michael Granitzer, Harald Kosch
Publication date:	05/26/2013
Journal:	Extended Semantic Web Conference
Pages:	268-270
Publisher:	Springer Berlin Heidelberg
Description:	A crucial task in a researchers’ daily work is the analysis of primary research data to estimate the evolution of certain fields or technologies, e.g. tables in publications or tabular benchmark results. Due to a lack of comparability and reliability of published primary research data, this becomes more and more time-consuming leading to contradicting facts, as has been shown for ad-hoc retrieval [1]. The CODE project [2] aims at contributing to a Linked Science Data Cloud by integrating unstructured research information with semantically represented research data. Through crowdsourcing techniques, data centric tasks like data extraction, integration and analysis in combination with sustainable data marketplace concepts will establish a sustainable, high-impact ecosystem.
Cited by:	7
Resources:	[PDF] from github.io

:view less

Models of human navigation in information networks based on decentralized search

2013

Denis Helic, Markus Strohmaier, Michael Granitzer, Reinhold Scherer
Proceedings of the 24th ACM conference on hypertext and social media, 89-98
:read more

Authors:	Denis Helic, Markus Strohmaier, Michael Granitzer, Reinhold Scherer
Publication date:	05/01/2013
Journal:	Proceedings of the 24th ACM conference on hypertext and social media
Pages:	89-98
Description:	Models of human navigation play an important role for understanding and facilitating user behavior in hypertext systems. In this paper, we conduct a series of principled experiments with decentralized search - an established model of human navigation in social networks - and study its applicability to information networks. We apply several variations of decentralized search to model human navigation in information networks and we evaluate the outcome in a series of experiments. In these experiments, we study the validity of decentralized search by comparing it with human navigational paths from an actual information network - Wikipedia. We find that (i) navigation in social networks appears to differ from human navigation in information networks in interesting ways and (ii) in order to apply decentralized search to information networks, stochastic adaptations are required. Our work illuminates a way towards using …
Cited by:	50
Resources:	[PDF] from acm.org

:view less

A survey on socio-semantic information retrieval

2013

Karin Schoefegger, Tanel Tammet, Michael Granitzer
Computer Science Review 8, 25-46
:read more

Authors:	Karin Schoefegger, Tanel Tammet, Michael Granitzer
Publication date:	05/01/2013
Journal:	Computer Science Review
Volume:	8
Pages:	25-46
Publisher:	Elsevier
Description:	The rise of the Social Web and advances in the Semantic Web provides unprecedented possibilities for the development of novel methods to enhance the information retrieval (IR) process by including varying degrees of semantics. We shed light on the corresponding notion of semantically-enhanced information retrieval by presenting state-of-the art techniques in related research areas. We describe techniques based on the main processes of a typical IR workflow and map them onto three main types of semantics, which vary from formal semantic knowledge representations and content-based semantics to social semantics emerging through usage and user interactions.
Cited by:	20
Resources:	[PDF] from academia.edu

:view less

I-KNOW 2013: Proceedings of the 13th International Conference on Knowledge Management and Knowledge Technologies: 4-6 September 2013, Graz, Austria

2013

Stefanie Lindstaedt, Michael Granitzer
:read more

Authors:	Stefanie Lindstaedt, Michael Granitzer
Publication date:	01/01/2013
Publisher:	Association for Computing Machinery

:view less

A Tabular Interface for the Semantic Web

2013

Patrick Höfler, Michael Granitzer, Vedran Sabol, Stefanie Lindstaedt
Proceedings of European Semantic Web Conference 2013 on Semantics and Big Data (ESWC 20…
:read more

Authors:	Patrick Höfler, Michael Granitzer, Vedran Sabol, Stefanie Lindstaedt
Publication date:	01/01/2013
Journal:	Proceedings of European Semantic Web Conference 2013 on Semantics and Big Data (ESWC 2013)
Pages:	1-1
Publisher:	.
Description:	A Tabular Interface for the Semantic Web - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Inhalt suchen unter Technische Universität Graz Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte A Tabular Interface for the Semantic Web Patrick Höfler , Michael Granitzer , Vedran Sabol , Stefanie Lindstaedt Know Center Research GmbH (98770) Publikation: Beitrag in Buch/Bericht/Konferenzband › Beitrag in einem Konferenzband › Begutachtung Übersicht Originalsprache englisch Titel Proceedings of European Semantic Web Conference 2013 on Semantics and Big Data (ESWC 2013) Herausgeber (Verlag) . Seiten 1-1 Publikationsstatus Veröffentlicht - 2013 Veranstaltung Extended Semantic Web Conference - Montpellier, …

:view less

D1. 1–First Conceptual Architecture and Requirements Definition

2013

Michael Granitzer, Christin Seifert
:read more

Authors:	Michael Granitzer, Christin Seifert
Publication date:	01/01/2013
Description:	D1.1 -- First Conceptual Architecture and Requirements Definition - University of Twente Research Information Skip to main navigation Skip to search Skip to main content University of Twente Research Information Home University of Twente Research Information Logo Home Profiles Research units Projects Research output Datasets Activities Prizes Press/Media Search by expertise, name or affiliation D1.1 -- First Conceptual Architecture and Requirements Definition Michael Granitzer, Christin Seifert Research output: Working paper Overview Original language English Publication status Published - 2013 Cite this APA Author BIBTEX Harvard Standard RIS Vancouver Powered by Pure, Scopus & Elsevier Fingerprint Engine™ All content on this site: Copyright © 2025 University of Twente Research Information, its licensors, and contributors. All rights are reserved, including those for text and data mining, AI training, …

:view less

D6. 1–Policy Model for Privacy Preservation and Feasibility Report

2013

Sonia Ben Mokhtar, Nadia Bennani, Lionel Brunie, Thomas Cerqueus, Elöd Egyed-Zsigmond, …
:read more

Authors:	Sonia Ben Mokhtar, Nadia Bennani, Lionel Brunie, Thomas Cerqueus, Elöd Egyed-Zsigmond, Omar Hasan, Albin Petit, Pierre-Edouard Portier, Michael Granitzer, Christin Seifert, Jörg Schlötterer
Publication date:	01/01/2013
Description:	D6.1 -- Policy Model for Privacy Preservation and Feasibility Report - University of Twente Research Information Skip to main navigation Skip to search Skip to main content University of Twente Research Information Home University of Twente Research Information Logo Home Profiles Research units Projects Research output Datasets Activities Prizes Press/Media Search by expertise, name or affiliation D6.1 -- Policy Model for Privacy Preservation and Feasibility Report Sonia Ben Mokhtar, Nadia Bennani, Lionel Brunie, Thomas Cerqueus, Elöd Egyed-Zsigmond, Omar Hasan, Albin Petit, Pierre-Edouard Portier, Michael Granitzer, Christin Seifert, Jörg Schlötterer Research output: Working paper Overview Original language English Publication status Published - 2013 Cite this APA Author BIBTEX Harvard Standard RIS Vancouver Powered by Pure, Scopus & Elsevier Fingerprint Engine™ All content on this site: …

:view less

Integrating User Preferences Into Distance Metrics.

2013

Mark Kröll, Vedran Sabol, Roman Kern, Michael Granitzer
LWA, 159-162
:read more

Authors:	Mark Kröll, Vedran Sabol, Roman Kern, Michael Granitzer
Publication date:	01/01/2013
Journal:	LWA
Pages:	159-162
Description:	Many practical scenarios such as search, classification or clustering benefit from better understanding their users, for instance, to deliver more relevant search results. Instead of committing ourselves to a specific field of research, eg by generating user profiles to enhance information retrieval, we seek to incorporate user preferences into the distance metric itself which lies at the heart of many algorithms including Information Retrieval and Machine Learning. The two approaches we explore in this paper allow users to directly convey their preferences in an intuitive way. The first approach adheres to the idea that just stating whether two documents are similar or not is more intuitive for a user than, for instance, assigning them to a broad spectrum of topics. The second approach seeks to take into account a user’s mental construct of the world being provided with a user-specific concept hierarchy. To evaluate our two approaches, we perform a text classification task. In the classification setting we use the Reuters RCV1 corpus to simulate user preferences. Our results indicate the principal feasibility of these two approaches and encourage further investigations.
Cited by:	1
Resources:	[PDF] from uni-bamberg.de

:view less

Towards disambiguating web tables

2013

Stefan Zwicklbauer, Christoph Einsiedler, Michael Granitzer, Christin Seifert
12th International Semantic Web Conference, ISWC 2013, 205-208
:read more

Authors:	Stefan Zwicklbauer, Christoph Einsiedler, Michael Granitzer, Christin Seifert
Publication date:	01/01/2013
Journal:	12th International Semantic Web Conference, ISWC 2013
Pages:	205-208
Publisher:	CEUR
Description:	Web tables comprise a rich source of factual information. However, without semantic annotation of the tables’ content the information is not usable for automatic integration and search. We propose a methodology to annotate table headers with semantic type information based on the content of column’s cells. In our experiments on 50 tables we achieved an F1 value of 0.55, where the accuracy greatly varies depending on the used ontology. Moreover, we found that for 94% of maximal F1 score only 20 cells (37%) need to be considered on average. Results suggest that for table disambiguation the choice of the ontology needs to be considered and the data input size can be reduced.
Cited by:	47
Resources:	[PDF] from ceur-ws.org

:view less

Text representation for efficient document annotation

2013

Christin Seifert, Eva Ulbrich, Roman Kern, Michael Granitzer
Journal of universal computer science 19 (3), 383-405
:read more

Authors:	Christin Seifert, Eva Ulbrich, Roman Kern, Michael Granitzer
Publication date:	01/01/2013
Journal:	Journal of universal computer science
Volume:	19
Issue:	3
Pages:	383-405
Publisher:	Technische Universitat Graz from Austria
Description:	In text classification the amount and quality of training data is crucial for the performance of the classifier. The generation of training data is done by human labellers-a tedious and time-consuming work. To reduce the labelling time for single documents we propose to use condensed representations of text documents instead of the full-text document. These condensed representations are key sentences and key phrases and can be generated in a fully unsupervised way. We extended and evaluated the TextRank algorithm to automatically extract key sentences and key phrases. For representing key phrases we propose a layout similar to a tag cloud. In a user study with 37 participants we evaluated whether document labelling with these condensed representations can be done faster and equally accurate by the human labellers. Our evaluation shows that the users labelled tag clouds twice as fast and as accurately as full-text documents. While further investigations for different classification tasks are necessary, this insight could potentially reduce costs for the labelling process of text documents.
Cited by:	9
Resources:	[PDF] from utwente.nl

:view less

Trends in semantic and digital media technologies

2013

Marcin Grzegorzek, Michael Granitzer, Stefan Rüger, Michael Sintek, Thierry Declerck, M…
Multimedia Tools and Applications 62 (2), 311-318
:read more

Authors:	Marcin Grzegorzek, Michael Granitzer, Stefan Rüger, Michael Sintek, Thierry Declerck, Massimo Romanelli
Publication date:	01/01/2013
Journal:	Multimedia Tools and Applications
Volume:	62
Issue:	2
Pages:	311-318
Publisher:	Springer US
Description:	The objective of this special issue is to report on recent trends in digital and media technologies responding to the challenges of managing and accessing multimedia (images, audio, video, 3D/4D material, etc.). In a highly selective review procedure we accepted contributions describing recent work that aims at narrowing the large disparity between the low-level multimedia descriptors and the richness of subjectivity of semantics in user queries and human interpretations of audiovisual data. The articles in this special issue can be grouped into following categories: Multimedia Analysis (Section 1), Multimedia Ontologies and Data Integration (Section 2), and Social Media and Retrieval (Section 3).
Cited by:	2
Resources:	[HTML] from springer.com

:view less

Visual ontology alignment system-an evaluation

2012

Vedran Sabol, Weng Onn Kow, Manuela Rauch, Eva Ulbrich, Christin Seifert, Michael Grani…
SIGRAD 2012: Interactive Visual Analysis of Data, 9-18
:read more

Authors:	Vedran Sabol, Weng Onn Kow, Manuela Rauch, Eva Ulbrich, Christin Seifert, Michael Granitzer, Dickson Lukose
Publication date:	11/01/2012
Journal:	SIGRAD 2012: Interactive Visual Analysis of Data
Pages:	9-18
Publisher:	Linkoping University Electronic Press
Description:	Ontology alignment is the process of mapping related concepts from different ontologies. A lot of research effort has been invested in development of algorithmic methods supporting automatic discovery of mappings between ontological concepts. However, automatic alignment remains potentially prone to errors especially with large real-world ontologies, demanding intervention of domain experts. We therefore created a semi-automatic tool including algorithmic alignment methods and an interactive visual interface. Visualisation components included in the interface support experts in navigating the concept space and reviewing the automatically generated mapping suggestions. An experiment with 15 test users was performed to evaluate whether, and in which cases the use of visualisation is beneficial compared to a user interface employing standard GUI widgets. The results indicate that users typically executed tasks slightly faster with an interface using standard widgets, but an interface which includes a visualisation component providing overview, filter and narrowing-down functionality achieved higher rates of successful task completion.
Cited by:	4
Resources:	[PDF] from utwente.nl

:view less

9th International Workshop on Text-Based Information Retrieval-TIR 2012-Preface

2012

Benno Stein, Michael Granitzer
2012 23rd International Workshop on Database and Expert Systems Applications, xx-xx
:read more

Authors:	Benno Stein, Michael Granitzer
Publication date:	09/03/2012
Journal:	2012 23rd International Workshop on Database and Expert Systems Applications
Pages:	xx-xx
Publisher:	IEEE
Description:	The workshop is being held for the ninth time. In the past, it was always characterized by a stimulating atmosphere, and it attracted high-quality contributions from all over the world. The TIR-12 Program Committee did a great job and provided us with detailed and professional reviews.

:view less

Dynamic topography information landscapes–an incremental approach to visual knowledge discovery

2012

Kamran Ali Ahmad Syed, Mark Kröll, Vedran Sabol, Arno Scharl, Stefan Gindl, Michael Gra…
International Conference on Data Warehousing and Knowledge Discovery, 352-363
:read more

Authors:	Kamran Ali Ahmad Syed, Mark Kröll, Vedran Sabol, Arno Scharl, Stefan Gindl, Michael Granitzer, Albert Weichselbraun
Publication date:	09/03/2012
Journal:	International Conference on Data Warehousing and Knowledge Discovery
Pages:	352-363
Publisher:	Springer Berlin Heidelberg
Description:	Incrementally computed information landscapes are an effective means to visualize longitudinal changes in large document repositories. Resembling tectonic processes in the natural world, dynamic rendering reflects both long-term trends and short-term fluctuations in such repositories. To visualize the rise and decay of topics, the mapping algorithm elevates and lowers related sets of concentric contour lines. Addressing the growing number of documents to be processed by state-of-the-art knowledge discovery applications, we introduce an incremental, scalable approach for generating such landscapes. The processing pipeline includes a number of sequential tasks, from crawling, filtering and pre-processing Web content to projecting, labeling and rendering the aggregated information. Incremental processing steps are localized in the projection stage consisting of document clustering, cluster force-directed …
Cited by:	8
Resources:	[HTML] from academia.edu

:view less

Teambeam-meta-data extraction from scientific literature

2012

Roman Kern, Kris Jack, Maya Hristakeva, Michael Granitzer
D-Lib Magazine 18 (7), 1
:read more

Authors:	Roman Kern, Kris Jack, Maya Hristakeva, Michael Granitzer
Publication date:	07/01/2012
Journal:	D-Lib Magazine
Volume:	18
Issue:	7
Pages:	1
Publisher:	Corporation for National Research Initiatives
Description:	An important aspect of the work of researchers as well as librarians is to manage collections of scientific literature. Social research networks, such as Mendeley and CiteULike, provide services that support this task. Meta-data plays an important role in providing services to retrieve and organise the articles. In such settings, meta-data is rarely explicitly provided, leading to the need for automatically extracting this valuable information. The TeamBeam algorithm analyses a scientific article and extracts structured meta-data, such as the title, journal name and abstract, as well as information about the article's authors (eg names, e-mail addresses, affiliations). The input of the algorithm is a set of blocks generated from the article text. A classification algorithm, which takes the sequence of the input into account, is then applied in two consecutive phases. In the evaluation of the algorithm, its performance is compared against two heuristics and three existing meta-data extraction systems. Three different data sets with varying characteristics are used to assess the quality of the extraction results. TeamBeam performs well under testing and compares favourably with existing approaches.
Cited by:	47
Resources:	[HTML] from dlib.org

:view less

Learning user characteristics from social tagging behavior

2012

Karin Schöfegger, Christian Körner, Philipp Singer, Michael Granitzer
Proceedings of the 23rd ACM conference on Hypertext and social media, 207-212
:read more

Authors:	Karin Schöfegger, Christian Körner, Philipp Singer, Michael Granitzer
Publication date:	06/25/2012
Journal:	Proceedings of the 23rd ACM conference on Hypertext and social media
Pages:	207-212
Description:	In social tagging systems the tagging activities of users leave a huge amount of implicit information about them. The users choose tags for the resources they annotate based on their interests, background knowledge, personal opinion and other criteria. Whilst existing research in mining social tagging data mostly focused on gaining a deeper understanding of the user's interests and the emerging structures in those systems, little work has yet been done to use the rich implicit information in tagging activities to unveil to what degree users' tags convey information about their background. The automatic inference of user background information can be used to complete user profiles which in turn supports various recommendation mechanisms. This work illustrates the application of supervised learning mechanisms to analyze a large online corpus of tagged academic literature for extraction of user characteristics from …
Cited by:	10
Resources:	[PDF] from acm.org

:view less

Navigational efficiency of broad vs. narrow folksonomies

2012

Denis Helic, Christian Körner, Michael Granitzer, Markus Strohmaier, Christoph Trattner
Proceedings of the 23rd ACM conference on Hypertext and social media, 63-72
:read more

Authors:	Denis Helic, Christian Körner, Michael Granitzer, Markus Strohmaier, Christoph Trattner
Publication date:	06/25/2012
Journal:	Proceedings of the 23rd ACM conference on Hypertext and social media
Pages:	63-72
Description:	Although many social tagging systems share a common tripartite graph structure, the collaborative processes that are generating these structures can differ significantly. For example, while resources on Delicious are usually tagged by all users who bookmark the web page cnn.com, photos on Flickr are usually tagged just by a single user who uploads the photo. In the literature, this distinction has been described as a distinction between broad vs. narrow folksonomies. This paper sets out to explore navigational differences between broad and narrow folksonomies in social hypertextual systems. We study both kinds of folksonomies on a dataset provided by Mendeley - a collaborative platform where users can annotate and organize scientific articles with tags. Our experiments suggest that broad folksonomies are more useful for navigation, and that the collaborative processes that are generating folksonomies matter …
Cited by:	34
Resources:	[PDF] from acm.org

:view less

On generating large-scale ground truth datasets for the deduplication of bibliographic records

2012

James A Hammerton, Michael Granitzer, Dan Harvey, Maya Hristakeva, Kris Jack
Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semanti…
:read more

Authors:	James A Hammerton, Michael Granitzer, Dan Harvey, Maya Hristakeva, Kris Jack
Publication date:	06/13/2012
Journal:	Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics
Pages:	1-12
Description:	Mendeley's crowd-sourced catalogue of research papers forms the basis of features such as the ability to search for papers, finding papers related to one currently being viewed and personalised recommendations. In order to generate this catalogue it is necessary to deduplicate the records uploaded from users' libraries and imported from external sources such as PubMed and arXiv. This task has been achieved at Mendeley via an automated system. However the quality of the deduplication needs to be improved. "Ground truth" data sets are thus needed for evaluating the system's performance but existing datasets are very small. In this paper, the problem of generating large scale data sets from Mendeley's database is tackled. An approach based purely on random sampling produced very easy data sets so approaches that focus on more difficult examples were explored. We found that selecting duplicates and …
Cited by:	22
Resources:	[PDF] from acm.org

:view less

Information retrieval and deduplication for tourism recommender sightsplanner

2012

Ago Luberg, Michael Granitzer, Honghan Wu, Priit Järv, Tanel Tammet
Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semanti…
:read more

Authors:	Ago Luberg, Michael Granitzer, Honghan Wu, Priit Järv, Tanel Tammet
Publication date:	06/13/2012
Journal:	Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics
Pages:	1-11
Description:	This paper is about scraping web pages for tourism objects and resolving duplicates for a tourism recommender system Sightsplanner. Gathering information from different web portals, we end up having several versions of the same object in our database. It is very important that we can find out which objects are duplicates and merge those. Only unique objects are presented to the end user. The main focus of this paper is therefore on deduplication problem. We have implemented a duplication detection system and tuned the parameters manually to get up to 85% accuracy. In this paper we present a machine learning setup which we used to improve deduplication accuracy of tourism attractions by 13 percentage points to achieve 98% accuracy. All the steps in the process are presented along with problems we tackled.
Cited by:	9
Resources:	[PDF] from acm.org

:view less

A comparison of layout based bibliographic metadata extraction techniques

2012

Michael Granitzer, Maya Hristakeva, Robert Knight, Kris Jack, Roman Kern
Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semanti…
:read more

Authors:	Michael Granitzer, Maya Hristakeva, Robert Knight, Kris Jack, Roman Kern
Publication date:	06/13/2012
Journal:	Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics
Pages:	1-8
Description:	Social research networks such as Mendeley and CiteULike offer various services for collaboratively managing bibliographic metadata. Compared with traditional libraries, metadata quality is of crucial importance in order to create a crowdsourced bibliographic catalog for search and browsing. Artifacts, in particular PDFs which are managed by the users of the social research networks, become one important metadata source and the starting point for creating a homogeneous, high quality, bibliographic catalog. Natural Language Processing and Information Extraction techniques have been employed to extract structured information from unstructured sources. However, given highly heterogeneous artifacts that cover a range of publication styles, stemming from different publication sources, and imperfect PDF processing tools, how accurate are metadata extraction methods in such real-world settings? This paper …
Cited by:	22
Resources:	[PDF] from acm.org

:view less

Unleashing semantics of research data

2012

Florian Stegmaier, Christin Seifert, Roman Kern, Patrick Höfler, Sebastian Bayerl, Mich…
Workshop on Big Data Benchmarks, 103-112
:read more

Authors:	Florian Stegmaier, Christin Seifert, Roman Kern, Patrick Höfler, Sebastian Bayerl, Michael Granitzer, Harald Kosch, Stefanie Lindstaedt, Belgin Mutlu, Vedran Sabol, Kai Schlegel, Stefan Zwicklbauer
Publication date:	05/08/2012
Journal:	Workshop on Big Data Benchmarks
Pages:	103-112
Publisher:	Springer Berlin Heidelberg
Description:	Research depends to a large degree on the availability and quality of primary research data, i.e., data generated through experiments and evaluations. While the Web in general and Linked Data in particular provide a platform and the necessary technologies for sharing, managing and utilizing research data, an ecosystem supporting those tasks is still missing. The vision of the CODE project is the establishment of a sophisticated ecosystem for Linked Data. Here, the extraction of knowledge encapsulated in scientific research paper along with its public release as Linked Data serves as the major use case. Further, Visual Analytics approaches empower end users to analyse, integrate and organize data. During these tasks, specific Big Data issues are present.
Cited by:	14
Resources:	[PDF] from academia.edu

:view less

Measuring the quality of web content using factual information

2012

Elisabeth Lex, Michael Voelske, Marcelo Errecalde, Edgardo Ferretti, Leticia Cagnina, C…
Proceedings of the 2nd joint WICOW/AIRWeb workshop on web quality, 7-10
:read more

Authors:	Elisabeth Lex, Michael Voelske, Marcelo Errecalde, Edgardo Ferretti, Leticia Cagnina, Christopher Horn, Benno Stein, Michael Granitzer
Publication date:	04/16/2012
Journal:	Proceedings of the 2nd joint WICOW/AIRWeb workshop on web quality
Pages:	7-10
Description:	Nowadays, many decisions are based on information found in the Web. For the most part, the disseminating sources are not certified, and hence an assessment of the quality and credibility of Web content became more important than ever. With factual density we present a simple statistical quality measure that is based on facts extracted from Web content using Open Information Extraction. In a first case study, we use this measure to identify featured/good articles in Wikipedia. We compare the factual density measure with word count, a measure that has successfully been applied to this task in the past. Our evaluation corroborates the good performance of word count in Wikipedia since featured/good articles are often longer than non-featured. However, for articles of similar lengths the word count measure fails while factual density can separate between them with an F-measure of 90.4%. We also investigate the use …
Cited by:	69
Resources:	[PDF] from acm.org

:view less

Harnessing user library statistics for research evaluation and knowledge domain visualization

2012

Peter Kraker, Christian Körner, Kris Jack, Michael Granitzer
Proceedings of the 21st international conference on world wide web, 1017-1024
:read more

Authors:	Peter Kraker, Christian Körner, Kris Jack, Michael Granitzer
Publication date:	04/16/2012
Journal:	Proceedings of the 21st international conference on world wide web
Pages:	1017-1024
Description:	Social reference management systems provide a wealth of information that can be used for the analysis of science. In this paper, we examine whether user library statistics can produce meaningful results with regards to science evaluation and knowledge domain visualization. We are conducting two empirical studies, using a sample of library data from Mendeley, the world's largest social reference management system. Based on the occurrence of references in users' libraries, we perform a large-scale impact factor analysis and an exploratory co-readership analysis. Our preliminary findings indicate that the analysis of user library statistics can produce accurate, timely, and content-rich results. We find that there is a significant relationship between the impact factor and the occurrence of references in libraries. Using a knowledge domain visualization based on co-occurrence measures, we are able to identify two …
Cited by:	23
Resources:	[PDF] from acm.org

:view less

A comparison of metadata extraction techniques for crowdsourced bibliographic metadata management

2012

Michael Granitzer, Maya Hristakeva, Kris Jack, Robert Knight
Proceedings of the 27th annual ACM symposium on applied computing, 962-964
:read more

Authors:	Michael Granitzer, Maya Hristakeva, Kris Jack, Robert Knight
Publication date:	03/26/2012
Journal:	Proceedings of the 27th annual ACM symposium on applied computing
Pages:	962-964
Description:	Social research networks such as Mendeley and CiteULike offer various services for collaboratively managing bibliographic metadata and uploading textual artifacts. One core problem thereby is the extraction of bibliographic metadata from the textual artifacts. Our work investiages the use of Conditional Random Fields and Support Vector Machines, implemented in two state-of-the-art real-world systems, namely ParsCit and the Mendeley Desktop, for automatically extracting bibliographic metadata. We compare the systems' accuracy on two newly created real-world data sets gathered from Mendeley and Linked-Open-Data repositories. Our analysis shows that two-stage SVMs provide reasonable performance in solving the challenge of metadata extraction from user-provided textual artifacts.
Cited by:	27
Resources:	[PDF] from acm.org

:view less

Overview and analysis of personal and social tagging context to construct user models

2012

Karin Schoefegger, Michael Granitzer
Proceedings of the 2nd Workshop on Context-awareness in Retrieval and Recommendation, 1…
:read more

Authors:	Karin Schoefegger, Michael Granitzer
Publication date:	02/14/2012
Journal:	Proceedings of the 2nd Workshop on Context-awareness in Retrieval and Recommendation
Pages:	14-21
Description:	The quality and user acceptance of personalized services such as personalized information retrieval and navigation or content recommendation depends depends besides the personalization mechanism on the quality, validity and accuracy of the employed user model. In literature a variety of user model construction methods based on tagging activity in social tagging systems (STS) are presented, relying on different user contexts, e.g., the personal or social context. But up to now there is neither a concise overview of existing construction methods available nor a deeper analysis and discussion of the differences between these models. Such an analysis would for example ease evaluation but also enable system designers to choose the most appropriate one. Our work tackles this problem by providing a short overview of state-of-the art user model construction methods which employ social tags. This is followed by a …
Cited by:	5
Resources:	[PDF] from acm.org

:view less

Mendeley’s open data for science and learning: a reply to the dataTEL challenge

2012

Kris Jack, Maya Hristakeva, Rosario Garcia de Zuniga, Michael Granitzer
International Journal of Technology Enhanced Learning 4 (1-2), 31-46
:read more

Authors:	Kris Jack, Maya Hristakeva, Rosario Garcia de Zuniga, Michael Granitzer
Publication date:	01/01/2012
Journal:	International Journal of Technology Enhanced Learning
Volume:	4
Issue:	1-2
Pages:	31-46
Publisher:	Inderscience Publishers
Description:	In this paper two open access data sources, provided by Mendeley Ltd., are described in detail. The first data source is a snapshot of 50,000 user libraries, enabling researchers to test collaborative filtering algorithms for scientific paper recommendation. The second data source is Mendeley’s API that provides access to 50 million research articles, crowdsourced from over 1.5 million users. This API allows researchers to build third party applications on top of real-time and large scale data. By providing data sources for the study of collaborative filtering, metadata extraction, deduplication and related research, Mendeley hopes to enable researchers in these domains to better collaborate and share knowledge, ultimately encouraging good research practices and advancing the state of scientific knowledge.
Cited by:	21
Resources:	[PDF] from academia.edu

:view less

Ontological model driven GUI development: User interface ontology approach

2011

Syed K Shahzad, Michael Granitzer, Denis Helic
2011 6th International Conference on Computer Sciences and Convergence Information Tech…
:read more

Authors:	Syed K Shahzad, Michael Granitzer, Denis Helic
Publication date:	11/29/2011
Journal:	2011 6th International Conference on Computer Sciences and Convergence Information Technology (ICCIT)
Pages:	214-218
Publisher:	IEEE
Description:	Ontology and Semantic Framework has become pervasive in computer science. It has huge impact at database, business logic and user interface for a range of computer applications. This framework is also being introduced, presented or plugged at user interfaces for various software and websites. However, establishment of structured and standardized ontological model based user interface development environment is still a challenge. This paper talks about the necessity of such an environment based on User Interface Ontology (UIO). To explore this phenomenon, this research focuses at the User Interface entities, their semantics, uses and relationships among them. The first part focuses on the development of User Interface Ontology. In the second step, this ontology is mapped to the domain ontology to construct a User Interface Model. Finally, the resulting model is quantified and instantiated for a user …
Cited by:	19

:view less

Future Internet and the Library world

2011

Michael Granitzer, Klaus Tochtermann
ZEITSCHRIFT FUR BIBLIOTHEKSWESEN UND BIBLIOGRAPHIE 58 (6), 324-328
:read more

Authors:	Michael Granitzer, Klaus Tochtermann
Publication date:	11/01/2011
Journal:	ZEITSCHRIFT FUR BIBLIOTHEKSWESEN UND BIBLIOGRAPHIE
Volume:	58
Issue:	6
Pages:	324-328
Publisher:	VITTORIO KLOSTERMANN GMBH

:view less

Realtime Ad Hoc Search in Twitter: Know-Center at TREC Microblog Track 2011.

2011

Christopher Horn, Oliver Pimas, Michael Granitzer, Elisabeth Lex, KC Graz
TREC
:read more

Authors:	Christopher Horn, Oliver Pimas, Michael Granitzer, Elisabeth Lex, KC Graz
Publication date:	11/01/2011
Journal:	TREC
Description:	In this paper, we outline our experiments carried out at the TREC Microblog Track 2011. Our system is based on a plain text index extracted from Tweets crawled from twitter. com. This index has been used to retrieve candidate Tweets for the given topics. The resulting Tweets were post-processed and then analyzed using three different approaches:(i) a burst detection approach,(ii) a hashtag analysis, and (iii) a Retweet analysis. Our experiments consisted of four runs: Firstly, a combination of the Lucene ranking with the burst detection, and secondly, a combination of the Lucene ranking, the burst detection, and the hashtag analysis. Thirdly, a combination of the Lucene ranking, the burst detection, the hashtag analysis, and the Retweet analysis, and fourthly, again a combination of the Lucene ranking with the burst detection but in this case with more sophisticated query language and post-processing. We achieved the best MAP values overall in the fourth run.
Cited by:	4
Resources:	[PDF] from nist.gov

:view less

A visual soa-based ontology alignment tool

2011

Weng Onn Kow, Vedran Sabol, Michael Granitzer, Wolfgang Kienrich, Dickson Lukose
Proceedings of the Sixth International Workshop on Ontology Matching (OM 2011) 10
:read more

Authors:	Weng Onn Kow, Vedran Sabol, Michael Granitzer, Wolfgang Kienrich, Dickson Lukose
Publication date:	10/24/2011
Journal:	Proceedings of the Sixth International Workshop on Ontology Matching (OM 2011)
Volume:	10
Description:	Ontology alignment is the process of matching related concepts from different ontologies. We propose a semi-automatic, visual approach which combines two algorithms for finding candidate alignments with visual navigation and analysis tools. The implementation is based on a Service-Oriented Architecture (SOA) to achieve scalability.
Cited by:	4
Resources:	[PDF] from ceur-ws.org

:view less

Word clouds for efficient document labeling

2011

Christin Seifert, Eva Ulbrich, Michael Granitzer
International Conference on Discovery Science, 292-306
:read more

Authors:	Christin Seifert, Eva Ulbrich, Michael Granitzer
Publication date:	10/05/2011
Journal:	International Conference on Discovery Science
Pages:	292-306
Publisher:	Springer Berlin Heidelberg
Description:	In text classification the amount and quality of training data is crucial for the performance of the classifier. The generation of training data is done by human labelers - a tedious and time-consuming work. We propose to use condensed representations of text documents instead of the full-text document to reduce the labeling time for single documents. These condensed representations are key sentences and key phrases and can be generated in a fully unsupervised way. The key phrases are presented in a layout similar to a tag cloud. In a user study with 37 participants we evaluated whether document labeling with these condensed representations can be done faster and equally accurate by the human labelers. Our evaluation shows that the users labeled word clouds twice as fast but as accurately as full-text documents. While further investigations for different classification tasks are necessary, this insight could …
Cited by:	17
Resources:	[PDF] from researchgate.net

:view less

Proceedings of the 11th International Conference on Knowledge Management and Knowledge Technologies

2011

Stefanie Lindstaedt, Michael Granitzer
:read more

Authors:	Stefanie Lindstaedt, Michael Granitzer
Publication date:	09/07/2011
Publisher:	ACM
Description:	The i-KNOW 2011, organized by Know-Center GmbH and Graz University of Technology, is the 11th conference of a very successful conference series on knowledge management and knowledge technologies. Since its inception in 2001, the International Conference on Knowledge Management and Knowledge Technologies (i-KNOW) has grown into an internationally visible and highly acknowledged event. Over 500 participants attend the conference every year.

:view less

Model selection strategies for author disambiguation

2011

Roman Kern, Mario Zechner, Michael Granitzer
2011 22nd International Workshop on Database and Expert Systems Applications, 155-159
:read more

Authors:	Roman Kern, Mario Zechner, Michael Granitzer
Publication date:	08/29/2011
Journal:	2011 22nd International Workshop on Database and Expert Systems Applications
Pages:	155-159
Publisher:	IEEE
Description:	Author disambiguation is a prerequisite for utilizing bibliographic metadata in citation analysis. Automatic disambiguation algorithms mostly rely on cluster-based disambiguation strategies for identifying unique authors given their names and publications. However, most approaches rely on knowing the correct number of unique authors a-priori, which is rarely the case in real world settings. In this publication we analyse cluster-based disambiguation strategies and develop a model selection method to estimate the number of distinct authors based on co-authorship networks. We show that, given clean textual features, the developed model selection method provides accurate guesses of the number of unique authors.
Cited by:	7
Resources:	[PDF] from webis.de

:view less

Semantic Multimedia: 5th International Conference on Semantic and Digital Media Technologies, SAMT 2010, Saarbrücken, Germany, December 1-3, 2010, Revised Selected Papers

2011

Thierry Declerck, Michael Granitzer, Marcin Grzegorzek, Massimo Romanelli, Stefan Rüger…
6725
:read more

Authors:	Thierry Declerck, Michael Granitzer, Marcin Grzegorzek, Massimo Romanelli, Stefan Rüger, Michael Sintek
Publication date:	08/19/2011
Volume:	6725
Publisher:	Springer Science & Business Media
Description:	This book constitutes the revised selected papers of the 5th International Conference on Semantics and Digital Media Technologies, SAMT 2010, held in Saarbrücken, Germany, in December 2010. As a result of a highly selective review procedure, 12 full papers and 4 short papers were accepted for publication. The contributions present novel approaches for managing, distributing and accessing large amounts of multimedia material. The topics covered include semantic search, analysis and retrieval of images, audio, video, 3D/4D material as well as of computer generated multimedia content. Also addressed are issues relating to semantic metadata management, semantic user interfaces, and semantics in visualization and computer graphics.
Cited by:	3

:view less

D2. 2–Performance report of coreference resolution in academic research papers

2011

Christin Seifert, Roman Kern, Michael Granitzer
:read more

Authors:	Christin Seifert, Roman Kern, Michael Granitzer
Publication date:	01/01/2011
Description:	D2.2 -- Performance report of coreference resolution in academic research papers - University of Twente Research Information Skip to main navigation Skip to search Skip to main content University of Twente Research Information Home University of Twente Research Information Logo Home Profiles Research units Projects Research output Datasets Activities Prizes Press/Media Search by expertise, name or affiliation D2.2 -- Performance report of coreference resolution in academic research papers Christin Seifert, Roman Kern, Michael Granitzer Research output: Working paper Overview Original language English Publication status Published - 2011 Externally published Yes Cite this APA Author BIBTEX Harvard Standard RIS Vancouver Powered by Pure, Scopus & Elsevier Fingerprint Engine™ All content on this site: Copyright © 2025 University of Twente Research Information, its licensors, and contributors. All …

:view less

Visual analyses on Linked Data–An opportunity for both fields

2011

Michael Granitzer, Vedran Sabol, Wolfgang Kienreich, Dickson Lukose, Kow Weng Onn
STI Semantic Summit, 0-0
:read more

Authors:	Michael Granitzer, Vedran Sabol, Wolfgang Kienreich, Dickson Lukose, Kow Weng Onn
Publication date:	01/01/2011
Journal:	STI Semantic Summit
Pages:	0-0
Publisher:	.
Description:	Visual Analyses on Linked Data – An Opportunity for both Fields - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Home Persons Publications Organisations Projects Activities Prizes Press/Media Search by expertise, name or affiliation Visual Analyses on Linked Data – An Opportunity for both Fields Michael Granitzer, Vedran Sabol, Wolfgang Kienreich, Dickson Lukose, Kow Weng Onn Know Center Research GmbH (98770) Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review Overview Original language English Title of host publication STI Semantic Summit Publisher . Pages 0-0 Publication status Published - 2011 Event STI Semantic Summit - Riga, Latvia Duration: 6 Jul 2011 → 8 Jul 2011 Conference Conference STI Semantic …
Cited by:	1

:view less

A visual soa-based ontology alignment tool

2011

Kow Weng Onn, Vedran Sabol, Michael Granitzer, Wolfgang Kienreich, Dickson Lukose
Sixth International Workshop on Ontology Matching (OM-2011)
:read more

Authors:	Kow Weng Onn, Vedran Sabol, Michael Granitzer, Wolfgang Kienreich, Dickson Lukose
Publication date:	01/01/2011
Journal:	Sixth International Workshop on Ontology Matching (OM-2011)
Description:	A Visual SOA-based Ontology Alignment Tool - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Search content at Graz University of Technology Home Persons Publications Organisations Projects Activities Prizes Press/Media A Visual SOA-based Ontology Alignment Tool Kow Weng Onn , Vedran Sabol , Michael Granitzer , Wolfgang Kienreich , Dickson Lukose Know Center Research GmbH (98770) Research output: Contribution to conference › Poster Overview Original language English Publication status Published - 2011 Event Sixth International Workshop on Ontology Matching (OM-2011) - Bonn, Germany Duration: 24 Oct 2011 → … Conference Conference Sixth International Workshop on Ontology Matching (OM-2011) City Bonn, Germany Period 24/10/11 → … Fields of …
Cited by:	1

:view less

Vote/veto meta-classifier for authorship identification notebook for PAN at CLEF 2011

2011

Roman Kern, Christin Seifert, Mario Zechner, Michael Granitzer
CLEF (Cross-Language Evaluation Forum) Conference on Multilingual and Multimodal Inform…
:read more

Authors:	Roman Kern, Christin Seifert, Mario Zechner, Michael Granitzer
Publication date:	01/01/2011
Journal:	CLEF (Cross-Language Evaluation Forum) Conference on Multilingual and Multimodal Information Access Evaluation 2011
Publisher:	CEUR
Description:	For the PAN 2011 authorship identification challenge we have devel-oped a system based on a meta-classifier which selectively uses the results of multiple base classifiers. In addition we also performed feature engineering based on the given domain of e-mails. We present our system as well as results on the evaluation dataset. Our system performed second and third best in the authorship attribution task on the large data sets, and ranked middle for the small data set in the attribution task and in the verification task.
Cited by:	19
Resources:	[PDF] from utwente.nl

:view less

Visualizing text classification models with Voronoi word clouds

2011

Christin Seifert, Wolfgang Kienreich, Michael Granitzer
:read more

Authors:	Christin Seifert, Wolfgang Kienreich, Michael Granitzer
Publication date:	01/01/2011
Description:	Visualizing Text Classification Models with Voronoi Word Clouds - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Visualizing Text Classification Models with Voronoi Word Clouds Christin Seifert, Wolfgang Kienreich, Michael Granitzer Know Center Research GmbH (98770) Publikation: Konferenzbeitrag › Poster Übersicht Originalsprache englisch Publikationsstatus Veröffentlicht - 2011 Veranstaltung 15th International Conference on Information Visualisation: IV 2011 - London, Großbritannien / Vereinigtes Königreich Dauer: 13 Juli 2011 → 15 Juli 2011 Konferenz Konferenz 15th International Conference on Information …
Cited by:	4

:view less

Web 2.0: Applications and Mechanisms J. UCS Special Issue

2011

Michael Granitzer, Stefanie N Lindstaedt
Journal of Universal Computer Science 17 (4), 513-514
:read more

Authors:	Michael Granitzer, Stefanie N Lindstaedt
Publication date:	01/01/2011
Journal:	Journal of Universal Computer Science
Volume:	17
Issue:	4
Pages:	513-514
Description:	This special issue features extended papers from the International Conference on Knowledge Management and Knowledge Technologies 2010 (I-KNOW 2010) held in Graz, Austria, September, 1-3, 2010. It comprises seven articles discussing Web 2.0 applications as well as technical mechanisms. The first three papers are concerned with Web 2.0 applications in different settings. The first paper is about the convergence of knowledge management and Enterprise 2.0 and describes the possibilities for an overarching exchange and transfer of knowledge on the basis of a case study at T-Systems Multimedia Solutions in which the decentralised development of ideas, collaboration and assistance to performing responsibilities played a central role. On this basis a regulatory framework as template for systemisation/definition of Enterprise 2.0 activities was developed. The next paper is about a web-based application at …

:view less

Who Tweets: Detecting user types and Tweet quality using supervised classification

2011

Christopher Horn, Elisabeth Lex, Michael Granitzer
IADIS International Conference Interfaces and Human Computer Interaction 2011, Part of …
:read more

Authors:	Christopher Horn, Elisabeth Lex, Michael Granitzer
Publication date:	01/01/2011
Journal:	IADIS International Conference Interfaces and Human Computer Interaction 2011, Part of the IADIS Multi Conference on Computer Science and Information Systems 2011: MCCSIS 2011
Pages:	153-157
Description:	Who Tweets: Detecting user types and Tweet quality using supervised classification - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Search content at Graz University of Technology Home Persons Publications Organisations Projects Activities Prizes Press/Media Who Tweets: Detecting user types and Tweet quality using supervised classification C. Horn , E. Lex , M. Granitzer Know Center Research GmbH (98770) Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review Overview Original language English Title of host publication Proceedings of the IADIS European Conference on Data Mining 2011 Pages 153-157 Publication status Published - 2011 Event IADIS International Conference Interfaces and Human Computer Interaction 2011, …

:view less

Knowledge Work: Knowledge Worker Productivity, Collaboration and User Support J. UCS Special Issue

2011

Michael Granitzer, Stefanie Lindstaedt
Journal of Universal Computer Science 17 (10), 1365-1366
:read more

Authors:	Michael Granitzer, Stefanie Lindstaedt
Publication date:	01/01/2011
Journal:	Journal of Universal Computer Science
Volume:	17
Issue:	10
Pages:	1365-1366
Description:	This special issue features extended papers from the International Conference on Knowledge Management and Knowledge Technologies 2010 (I-KNOW 2010) held in Graz, Austria, September, 1-3, 2010. It comprises seven articles focussing on various aspects of knowledge work, namely knowledge worker productivity, collaboration and user support. The first two papers address the aspect of knowledge worker productivity. The first paper is mainly concerned with the measurement of knowledge worker productivity as regards experts in specific professional domains. The results of an empirical study in different industrial domains suggests that productivity itself is not useable as performance indicator but that it is composed of multiple indicators which additionally are of different importance for different stakeholders. The second paper also states that knowledge worker productivity-according to the authors composed …
Cited by:	2

:view less

Semantic Web: Theory and Applications J. UCS Special Issue

2011

Michael Granitzer, Stefanie Lindstaedt
Journal of Universal Computer Science 17 (7), 981-982
:read more

Authors:	Michael Granitzer, Stefanie Lindstaedt
Publication date:	01/01/2011
Journal:	Journal of Universal Computer Science
Volume:	17
Issue:	7
Pages:	981-982
Description:	This special issue features extended papers from the International Conference on Knowledge Management and Knowledge Technologies 2010 (I-KNOW 2010) held in Graz, Austria, September, 1-3, 2010. It comprises eight articles addressing theoretical as well as application oriented aspects of the Semantic Web. The first three papers pick up theoretical aspects of Semantic Web, namely RDF, ontology evaluation and ontology development. The first article picks up the problem of query languages for RDF which however do not support fuzzy queries. The authors introduce a new method that transforms subject-relation-object into activation patterns. This represents a common model as basis for sophisticated analysis methods such as semantic relation analysis or anomaly detection. The Activation Patterns concept is introduced and applied to an RDF representation of the well-known CIA World Factbook. The …
Cited by:	2

:view less

User-based active learning

2010

Christin Seifert, Michael Granitzer
2010 IEEE international conference on data mining workshops, 418-425
:read more

Authors:	Christin Seifert, Michael Granitzer
Publication date:	12/13/2010
Journal:	2010 IEEE international conference on data mining workshops
Pages:	418-425
Publisher:	IEEE
Description:	Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more intelligent input from the user may be beneficial. For instance, given only one example at a time users are hardly able to determine whether this example is an outlier or not. In this paper we propose user-based visually-supported active learning strategies that allow the user to do both, selecting and labeling examples given a trained classifier. While labeling is straightforward, selection takes place using a interactive visualization of the classifier's a-posteriori output probabilities. By simulating different user selection strategies we show, that user-based active learning outperforms uncertainty based sampling methods and yields a …
Cited by:	65
Resources:	[PDF] from researchgate.net

:view less

Combining BPM and social software: contradiction or chance?

2010

Selim Erol, Michael Granitzer, Simone Happ, Sami Jantunen, Ben Jennings, Paul Johanness…
Journal of software maintenance and evolution: research and practice 22 (6‐7), 449-476
:read more

Authors:	Selim Erol, Michael Granitzer, Simone Happ, Sami Jantunen, Ben Jennings, Paul Johannesson, Agnes Koschmider, Selmin Nurcan, Davide Rossi, Rainer Schmidt
Publication date:	10/01/2010
Journal:	Journal of software maintenance and evolution: research and practice
Volume:	22
Issue:	6‐7
Pages:	449-476
Publisher:	John Wiley & Sons, Ltd.
Description:	Social software has received much attention in the academia and industry due to many success stories. However, although social software is used widely for business support, its relationship with Business Process Management has not been analysed. The results of the workshop on Business Process Management and Social Software (BPMS2'08), as part of the International Conference on Business Process Management in Milano, show the manifold possibilities of combining concepts from Business Process Management and social software. Social software provides a better integration of all stakeholders into the business process life cycle and offers new possibilities for a more effective and flexible design of business processes. The modelling of business processes may profit particularly from using social software techniques by alleviating the integration of process knowledge from all stakeholders. In addition …
Cited by:	197
Resources:	[PDF] from academia.edu

:view less

German encyclopedia alignment based on information retrieval techniques

2010

Roman Kern, Michael Granitzer
International Conference on Theory and Practice of Digital Libraries, 315-326
:read more

Authors:	Roman Kern, Michael Granitzer
Publication date:	09/06/2010
Journal:	International Conference on Theory and Practice of Digital Libraries
Pages:	315-326
Publisher:	Springer Berlin Heidelberg
Description:	Collaboratively created online encyclopedias have become increasingly popular. Especially in terms of completeness they have begun to surpass their printed counterparts. Two German publishers of traditional encyclopedias have reacted to this challenge and decided to merge their corpora to create a single more complete encyclopedia. The crucial step in this merge process is the alignment of articles. We have developed a system to identify corresponding entries from different encyclopedic corpora. The base of our system is the alignment algorithm which incorporates various techniques developed in the field of information retrieval. We have evaluated the system on four real-world encyclopedias with a ground truth provided by domain experts. A combination of weighting and ranking techniques has been found to deliver a satisfying performance.
Cited by:	6
Resources:	[PDF] from researchgate.net

:view less

External and intrinsic plagiarism detection using a cross-lingual retrieval and segmentation system

2010

Markus Muhr, Roman Kern, Mario Zechner, Michael Granitzer
Notebook papers of CLEF 2010 LABs and workshops, 22
:read more

Authors:	Markus Muhr, Roman Kern, Mario Zechner, Michael Granitzer
Publication date:	09/01/2010
Journal:	Notebook papers of CLEF 2010 LABs and workshops
Pages:	22
Description:	We present our hybrid system for the PAN challenge at CLEF 2010. Our system performs plagiarism detection for translated and non-translated externally as well as intrinsically plagiarized document passages. Our external plagiarism detection approach is formulated as an information retrieval problem, using heuristic post processing to arrive at the final detection results. For the retrieval step, source documents are split into overlapping blocks which are indexed via a Lucene instance. Suspicious documents are similarly split into consecutive overlapping boolean queries which are performed on the Lucene index to retrieve an initial set of potentially plagiarized passages. For performance reasons queries might get rejected via a heuristic before actually being executed. Candidate hits gathered via the retrieval step are further post-processed by performing sequence analysis on the passages retrieved from the index with respect to the passages used for querying the index. By applying several merge heuristics bigger blocks are formed from matching sequences. German and Spanish source documents are first translated using word alignment on the Europarl corpus before entering the above detection process. For each word in a translated document several translations are produced. Intrinsic plagiarism detection is done by finding major changes in style measured via word suffixes after the documents have been partitioned by an linear text segmentation algorithm. Our approach lead us to the third overall rank with an overall score of 0.6948.
Cited by:	60
Resources:	[PDF] from webis.de

:view less

A comparison of stylometric and lexical features for web genre classification and emotion classification in blogs

2010

Elisabeth Lex, Andreas Juffinger, Michael Granitzer
2010 Workshops on Database and Expert Systems Applications, 10-14
:read more

Authors:	Elisabeth Lex, Andreas Juffinger, Michael Granitzer
Publication date:	08/30/2010
Journal:	2010 Workshops on Database and Expert Systems Applications
Pages:	10-14
Publisher:	IEEE
Description:	In the blogosphere, the amount of digital content is expanding and for search engines, new challenges have been imposed. Due to the changing information need, automatic methods are needed to support blog search users to filter information by different facets. In our work, we aim to support blog search with genre and facet information. Since we focus on the news genre, our approach is to classify blogs into news versus rest. Also, we assess the emotionality facet in news related blogs to enable users to identify people's feelings towards specific events. Our approach is to evaluate the performance of text classifiers with lexical and stylometric features to determine the best performing combination for our tasks. Our experiments on a subset of the TREC Blogs08 dataset reveal that classifiers trained on lexical features perform consistently better than classifiers trained on the best stylometric features.
Cited by:	13
Resources:	[PDF] from semanticscholar.org

:view less

Scalable recursive top-down hierarchical clustering approach with implicit model selection for textual data sets

2010

Markus Muhr, Vedran Sabol, Michael Granitzer
2010 Workshops on Database and Expert Systems Applications, 15-19
:read more

Authors:	Markus Muhr, Vedran Sabol, Michael Granitzer
Publication date:	08/30/2010
Journal:	2010 Workshops on Database and Expert Systems Applications
Pages:	15-19
Publisher:	IEEE
Description:	Automatic generation of taxonomies can be useful for a wide area of applications. In our application scenario a topical hierarchy should be constructed reasonably fast from a large document collection to aid browsing of the data set. The hierarchy should also be used by the InfoSky projection algorithm to create an information landscape visualization suitable for explorative navigation of the data. We developed an algorithm that applies a scalable, recursive, top-down clustering approach to generate a dynamic concept hierarchy. The algorithm recursively applies a workflow consisting of preprocessing, clustering, cluster labeling and projection into 2D space. Besides presenting and discussing the benefits of combining hierarchy browsing with visual exploration, we also investigate the clustering results achieved on a real world data set.
Cited by:	14
Resources:	[PDF] from webis.de

:view less

Ontology alignment—a survey with focus on visually supported semi-automatic techniques

2010

Michael Granitzer, Vedran Sabol, Kow Weng Onn, Dickson Lukose, Klaus Tochtermann
Future Internet 2 (3), 238-258
:read more

Authors:	Michael Granitzer, Vedran Sabol, Kow Weng Onn, Dickson Lukose, Klaus Tochtermann
Publication date:	08/04/2010
Journal:	Future Internet
Volume:	2
Issue:	3
Pages:	238-258
Publisher:	MDPI
Description:	Semantic technologies are of paramount importance to the future Internet. The reuse and integration of semantically described resources, such as data or services, necessitates the bringing of ontologies into mutual agreement. Ontology alignment deals with the discovery of correspondences between concepts and relations from different ontologies. Alignment provides the key ingredient to semantic interoperability. This paper gives an overview on the state of the art in the field of visually supported semi-automatic alignment techniques and presents recent trends and developments. Particular attention is given to user interfaces and visualization techniques supporting involvement of humans in the alignment process. We derive and summarize requirements for visual semi-automatic alignment systems, provide an overview of existing approaches, and discuss the possibilities for further improvements and future research.
Cited by:	115
Resources:	[HTML] from mdpi.com

:view less

Semantically Enhanced Software Documentation Processes

2010

Werner Klieber, Michael Granitzer, Mansuet Gaisbauer, Klaus Tochtermann
Serdica Journal of Computing 4 (2), 243-262
:read more

Authors:	Werner Klieber, Michael Granitzer, Mansuet Gaisbauer, Klaus Tochtermann
Publication date:	07/20/2010
Journal:	Serdica Journal of Computing
Volume:	4
Issue:	2
Pages:	243-262
Description:	High-quality software documentation is a substantial issue for understanding software systems. Shorter time-to-market software cycles increase the importance of automatism for keeping the documentation up to date. In this paper, we describe the automatic support of the software documentation process using semantic technologies. We introduce a software documentation ontology as an underlying knowledge base. The defined ontology is populated automatically by analysing source code, software documentation and code execution. Through selected results we demonstrate that the use of such semantic systems can support software documentation processes efficiently.
Resources:	[PDF] from bas.bg

:view less

Analysis of structural relationships for hierarchical cluster labeling

2010

Markus Muhr, Roman Kern, Michael Granitzer
Proceedings of the 33rd international ACM SIGIR conference on Research and development …
:read more

Authors:	Markus Muhr, Roman Kern, Michael Granitzer
Publication date:	07/19/2010
Journal:	Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval
Pages:	178-185
Description:	Cluster label quality is crucial for browsing topic hierarchies obtained via document clustering. Intuitively, the hierarchical structure should influence the labeling accuracy. However, most labeling algorithms ignore such structural properties and therefore, the impact of hierarchical structures on the labeling accuracy is yet unclear. In our work we integrate hierarchical information, i.e. sibling and parent-child relations, in the cluster labeling process. We adapt standard labeling approaches, namely Maximum Term Frequency, Jensen-Shannon Divergence, Chi Square Test, and Information Gain, to take use of those relationships and evaluate their impact on 4 different datasets, namely the Open Directory Project, Wikipedia, TREC Ohsumed and the CLEF IP European Patent dataset. We show, that hierarchical relationships can be exploited to increase labeling accuracy especially on high-level nodes.
Cited by:	38
Resources:	[PDF] from acm.org

:view less

Classifier hypothesis generation using visual analysis methods

2010

Christin Seifert, Vedran Sabol, Michael Granitzer
International Conference on Networked Digital Technologies, 98-111
:read more

Authors:	Christin Seifert, Vedran Sabol, Michael Granitzer
Publication date:	07/07/2010
Journal:	International Conference on Networked Digital Technologies
Pages:	98-111
Publisher:	Springer Berlin Heidelberg
Description:	Classifiers can be used to automatically dispatch the abundance of newly created documents to recipients interested in particular topics. Identification of adequate training examples is essential for classification performance, but it may prove to be a challenging task in large document repositories. We propose a classifier hypothesis generation method relying on automated analysis and information visualisation. In our approach visualisations are used to explore the document sets and to inspect the results of machine learning methods, allowing the user to assess the classifier performance and adapt the classifier by gradually refining the training set.
Cited by:	12

:view less

KCDC: Word sense induction by using grammatical dependencies and sentence phrase structure

2010

Roman Kern, Markus Muhr, Michael Granitzer
Proceedings of the 5th international workshop on semantic evaluation, 351-354
:read more

Authors:	Roman Kern, Markus Muhr, Michael Granitzer
Publication date:	07/01/2010
Journal:	Proceedings of the 5th international workshop on semantic evaluation
Pages:	351-354
Description:	Word sense induction and discrimination (WSID) identifies the senses of an ambiguous word and assigns instances of this word to one of these senses. We have build a WSID system that exploits syntactic and semantic features based on the results of a natural language parser component. To achieve high robustness and good generalization capabilities, we designed our system to work on a restricted, but grammatically rich set of features. Based on the results of the evaluations our system provides a promising performance and robustness.
Cited by:	14
Resources:	[PDF] from aclanthology.org

:view less

Objectivity classification in online media

2010

Elisabeth Lex, Andreas Juffinger, Michael Granitzer
Proceedings of the 21st ACM conference on Hypertext and hypermedia, 293-294
:read more

Authors:	Elisabeth Lex, Andreas Juffinger, Michael Granitzer
Publication date:	06/13/2010
Journal:	Proceedings of the 21st ACM conference on Hypertext and hypermedia
Pages:	293-294
Description:	In this work, we assess objectivity in online news media. We propose to use topic independent features and we show in a cross-domain experiment that with standard bag-of-word models, classifiers implicitly learn topics. Our experiments revealed that our methodology can be applied across different topics with consistent classification performance.
Cited by:	28
Resources:	[PDF] from acm.org

:view less

A hybrid system for German encyclopedia alignment

2010

Roman Kern, Christin Seifert, Michael Granitzer
International journal on digital libraries 11 (2), 75-89
:read more

Authors:	Roman Kern, Christin Seifert, Michael Granitzer
Publication date:	06/01/2010
Journal:	International journal on digital libraries
Volume:	11
Issue:	2
Pages:	75-89
Publisher:	Springer Berlin Heidelberg
Description:	Collaboratively created on-line encyclopedias have become increasingly popular. Especially in terms of completeness they have begun to surpass their printed counterparts. Two German publishers of traditional encyclopedias have reacted to this challenge and started an initiative to merge their corpora to create a single, more complete encyclopedia. The crucial step in this merging process is the alignment of articles. We have developed a two-step hybrid system to provide high-accurate alignments with low manual effort. First, we apply an information retrieval based, automatic alignment algorithm. Second, the articles with a low confidence score are revised using a manual alignment scheme carefully designed for quality assurance. Our evaluation shows that a combination of weighting and ranking techniques utilizing different facets of the encyclopedia articles allow to effectively reduce the number of necessary …
Cited by:	3

:view less

Stylometric features for emotion level classification in news related blogs

2010

Elisabeth Lex, Michael Granitzer, Markus Muhr, Andreas Juffinger
Adaptivity, Personalization and Fusion of Heterogeneous Information, 132-133
:read more

Authors:	Elisabeth Lex, Michael Granitzer, Markus Muhr, Andreas Juffinger
Publication date:	04/28/2010
Journal:	Adaptivity, Personalization and Fusion of Heterogeneous Information
Pages:	132-133
Description:	Breaking news and events are often posted in the blogosphere before they are published by any media agency. Therefore, the blogosphere is a valuable resource for news-related blog analysis. However, it is crucial to first sort out newsunrelated content like personal diaries or advertising blogs. Besides, there are different levels of emotionality or involvement which bias the news information to a certain extent. In our work, we evaluate topic-independent stylometric features to classify blogs into news versus rest and to assess the emotionality in these blogs. We apply several text classifiers to determine the best performing combination of features and algorithms. Our experiments revealed that with simple style features, blogs can be classified into news versus rest and their emotionality can be assessed with accuracy values of almost 80%.
Cited by:	7
Resources:	[PDF] from researchgate.net

:view less

Adaptive term weighting through stochastic optimization

2010

Michael Granitzer
International Conference on Intelligent Text Processing and Computational Linguistics, …
:read more

Authors:	Michael Granitzer
Publication date:	03/21/2010
Journal:	International Conference on Intelligent Text Processing and Computational Linguistics
Pages:	614-626
Publisher:	Springer Berlin Heidelberg
Description:	Term weighting strongly influences the performance of text mining and information retrieval approaches. Usually term weights are determined through statistical estimates based on static weighting schemes. Such static approaches lack the capability to generalize to different domains and different data sets. In this paper, we introduce an on-line learning method for adapting term weights in a supervised manner. Via stochastic optimization we determine a linear transformation of the term space to approximate expected similarity values among documents. We evaluate our approach on 18 standard text data sets and show that the performance improvement of a k-NN classifier ranges between 1% and 12% by using adaptive term weighting as preprocessing step. Further, we provide empirical evidence that our approach is efficient to cope with larger problems.
Cited by:	1
Resources:	[PDF] from researchgate.net

:view less

Editorial Preface: Special issue on context–aware and mobile multimedia databases and services

2010

Romulus Grigoras, Vincent Charvillat, Richard Chbeir, Harald Kosch, Werner Bailer, Susa…
Journal of Digital Information Management 8 (1)
:read more

Authors:	Romulus Grigoras, Vincent Charvillat, Richard Chbeir, Harald Kosch, Werner Bailer, Susanne Boll, Lionel Brunie, Yiwei Cao, Jaime Delgado, Michael Granitzer, Michael Hausenblas
Publication date:	02/01/2010
Journal:	Journal of Digital Information Management
Volume:	8
Issue:	1
Publisher:	Digital Information Research Foundation
Description:	It is expected in the near future that the number of created digital multimedia content will rise dramatically. Multimedia metadata is currently the only way to cope with problems like semantic-based information retrieval or organization of content, and provides means to specify adaptation and delivery constraints and rules. In parallel, the integration of accurate hand-held positioning within mobile phones, PDAs, portable computer and the like has brought locationbased services directly to individuals, transforming the way we work, travel or play. Therefore, the context has become an important element to be considered in mobile applications and scenarios. Its usage provides new hints for search, delivery and consumption to individual user’s situations.

:view less

Wissensmanagement in EBM

2010

Gerald Lirk, Michael Granitzer, Andreas Sönnichsen, Peter Kulczycki
EbM–ein Gewinn für die Arzt-Patient-Beziehung?. Forum Medizin 21 der Paracelsus Medizin…
:read more

Authors:	Gerald Lirk, Michael Granitzer, Andreas Sönnichsen, Peter Kulczycki
Publication date:	01/01/2010
Journal:	EbM–ein Gewinn für die Arzt-Patient-Beziehung?. Forum Medizin 21 der Paracelsus Medizinischen Privatuniversität & 11. EbM-Jahrestagung des Deutschen Netzwerks Evidenzbasierte Medizin
Pages:	0-0
Publisher:	.
Description:	Wissensmanagement in EBM - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Wissensmanagement in EBM Gerald Lirk, Michael Granitzer, Andreas Sönnichsen, Peter Kulczycki Publikation: Beitrag in Buch/Bericht/Konferenzband › Beitrag in einem Konferenzband › Begutachtung Übersicht Originalsprache deutsch Titel EbM – ein Gewinn für die Arzt-Patient-Beziehung?. Forum Medizin 21 der Paracelsus Medizinischen Privatuniversität & 11. EbM-Jahrestagung des Deutschen Netzwerks Evidenzbasierte Medizin Herausgeber (Verlag) . Seiten 0-0 Publikationsstatus Veröffentlicht - 2010 Dieses zitieren APA Standard Harvard …

:view less

Semantische Technologien: Stand der Forschung und Visionen

2010

Michael Granitzer, Wolfgang Kienreich
OCG: Internationales Rechtsinformatik Symposion (IRIS 10), 0-0
:read more

Authors:	Michael Granitzer, Wolfgang Kienreich
Publication date:	01/01/2010
Journal:	OCG: Internationales Rechtsinformatik Symposion (IRIS 10)
Pages:	0-0
Publisher:	.
Description:	Semantische Technologien: Stand der Forschung und Visionen - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Semantische Technologien: Stand der Forschung und Visionen Michael Granitzer, Wolfgang Kienreich Know Center Research GmbH (98770) Publikation: Beitrag in Buch/Bericht/Konferenzband › Beitrag in einem Konferenzband › Begutachtung Übersicht Originalsprache deutsch Titel OCG: Internationales Rechtsinformatik Symposion (IRIS 10) Herausgeber (Verlag) . Seiten 0-0 Publikationsstatus Veröffentlicht - 2010 Dieses zitieren APA Standard Harvard Vancouver Author BIBTEX RIS Angetrieben von Pure, Scopus …

:view less

Knowledge Relationship Discovery and Visually Enhanced Access for the Media Domain

2010

Michael Granitzer, Wolfgang Kienreich, Vedran Sabol, Elisabeth Lex
, 46-58
:read more

Authors:	Michael Granitzer, Wolfgang Kienreich, Vedran Sabol, Elisabeth Lex
Publication date:	01/01/2010
Pages:	46-58
Publisher:	Innsbruck University Press
Description:	Technological advances and paradigmatic changes in the utilization of the World Wide Web have transformed the information seeking strategies of media consumers and invalidated traditional business models of media providers. We discuss relevant aspects of this development and present a knowledge relationship discovery pipeline to address the requirements of media providers and media consumers. We also propose visually enhanced access methods to bridge the gap between complex media services and the information needs of the general public. We conclude that a combination of advanced processing methods and visualizations will enable media providers to take the step from content-centered to service-centered business models and, at the same time, will help media consumers to better satisfy their personal information needs.
Cited by:	1
Resources:	[PDF] from oapen.org

:view less

Ontological Framework Driven GUI Development

2010

Syed K Shahzad, Michael Granitzer
Proceedings of I-KNOW, 198-206
:read more

Authors:	Syed K Shahzad, Michael Granitzer
Publication date:	01/01/2010
Journal:	Proceedings of I-KNOW
Pages:	198-206
Description:	The user experience of any software or website consists of elements from the conceptual to the concrete level. These elements of user experience assist in the design and development of user interfaces. On the other hand, Ontologies provide a framework for computable representation of user interface elements and underlying data. This paper discusses strategies of introducing ontologies at different user interface layers adapted from user experience elements. These layers range from abstract levels (eg User needs/Application Objectives) to concrete levels (eg Application User Interface) in term of data representation. The proposed ontological framework enables device independent, semi-automated GUI construction which we will demonstrate at a personal information management example.
Cited by:	1
Resources:	[PDF] from researchgate.net

:view less

Efficient cross-domain classification of weblogs

2010

Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
International Journal of Intelligent Computing Research 1 (3), 69-76
:read more

Authors:	Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
Publication date:	01/01/2010
Journal:	International Journal of Intelligent Computing Research
Volume:	1
Issue:	3
Pages:	69-76
Publisher:	Infonomics Society
Description:	Text classification is one of the core applications in data mining due to the huge amount of uncategorized textual data available. Training a text classifier results in a classification model that reflects the characteristics of the domain it was learned on. However, if no training data is available, labeled data from a related but different domain might be exploited to perform cross-domain classification. In our work, we aim to accurately classify unlabeled weblogs into commonly agreed upon newspaper categories using labeled data from the news domain. The labeled news and the unlabeled blog corpus are highly dynamic and hourly growing with a topic drift, so the classification needs to be efficient. Our approach is to apply a fast novel centroid-based text classification algorithm, the Class-Feature-Centroid Classifier (CFC), to perform efficient cross-domain classification. Experiments showed that this algorithm achieves a comparable accuracy than k-Nearest Neighbour (k-NN) and Support Vector Machines (SVM), yet at linear time cost for training and classification. We investigate the classifier performance and generalization ability using a special visualization of classifiers. The benefit of our approach is that the linear time complexity enables us to efficiently generate an accurate classifier, reflecting the topic drift, several times per day on a huge dataset.
Cited by:	14
Resources:	[PDF] from utwente.nl

:view less

Semantic Web services: state of the art

2010

Markus Lanthaler, Michael Granitzer, Christian Gütl
Proceedings of the IADIS International Conference on Internet Technologies & Society (I…
:read more

Authors:	Markus Lanthaler, Michael Granitzer, Christian Gütl
Publication date:	01/01/2010
Journal:	Proceedings of the IADIS International Conference on Internet Technologies & Society (ITS 2010)
Pages:	107-114
Description:	Service-oriented architectures (SOA) built on Web services were a first attempt to streamline and automate business processes in order to increase productivity but the utopian promise of uniform service interface standards, metadata, and universal service registries, in the form of the SOAP, WSDL and UDDI standards has proven elusive. Furthermore, the RPC-oriented model of those traditional Web services is not Web-friendly. Thus more and more prominent Web service providers opted to expose their services based on the REST architectural style. Nevertheless there are still problems on formal describing, finding, and orchestrating RESTful services. While there are already a number of different approaches none so far has managed to break out of its academic confines. This paper focuses on an extensive survey comparing the existing state-of-the-art technologies for semantically annotated Web services as a first step towards a proposal designed specifically for RESTful services.
Cited by:	18
Resources:	[PDF] from markus-lanthaler.com

:view less

A competitive learning approach to instance selection for support vector machines

2009

Mario Zechner, Michael Granitzer
International Conference on Knowledge Science, Engineering and Management, 146-157
:read more

Authors:	Mario Zechner, Michael Granitzer
Publication date:	11/25/2009
Journal:	International Conference on Knowledge Science, Engineering and Management
Pages:	146-157
Publisher:	Springer Berlin Heidelberg
Description:	Support Vector Machines (SVM) have been applied successfully in a wide variety of fields in the last decade. The SVM problem is formulated as a convex objective function subject to box constraints that needs to be maximized, a quadratic programming (QP) problem. In order to solve the QP problem on larger data sets specialized algorithms and heuristics are required. In this paper we present a new data-squashing method for selecting training instances in support vector learning. Inspired by the growing neural gas algorithm and learning vector quantization we introduce a new, parameter robust neural gas variant to retrieve an initial approximation of the training set containing only those samples that will likely become support vectors in the final classifier. This first approximation is refined in the border areas, defined by neighboring neurons of different classes, yielding the final training set. We evaluate our …
Cited by:	5
Resources:	[PDF] from researchgate.net

:view less

Designing user interfaces through ontological user model: functional programming approach

2009

Syed K Shahzad, Michael Granitzer, Klause Tochterman
2009 Fourth International Conference on Computer Sciences and Convergence Information T…
:read more

Authors:	Syed K Shahzad, Michael Granitzer, Klause Tochterman
Publication date:	11/24/2009
Journal:	2009 Fourth International Conference on Computer Sciences and Convergence Information Technology
Pages:	99-104
Publisher:	IEEE
Description:	Semantics and Ontological framework defines standards for concepts presented through computational models. This framework can also provide standards for the concept representation at User Interface level. A user model as the base of user interface can be built using formal ontologies. Language supporting Higher Order Functions can be used to build Graphical User Interfaces by mapping the user defined model with typed graphical user interface controls. This paper discusses a strategy to present the standard concepts of computational model to user interface through ontological user model.
Cited by:	15

:view less

Cross-domain classification: Trade-off between complexity and accuracy

2009

Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
2009 International Conference for Internet Technology and Secured Transactions,(ICITST)…
:read more

Authors:	Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
Publication date:	11/09/2009
Journal:	2009 International Conference for Internet Technology and Secured Transactions,(ICITST)
Pages:	1-6
Publisher:	IEEE
Description:	Text classification is one of the core applications in data mining due to the huge amount of not categorized digital data available. Training a text classifier generates a model that reflects the characteristics of the domain. However, if no training data is available, labeled data from a related but different domain might be exploited to perform cross-domain classification. In our work, we aim to accurately classify unlabeled blogs into commonly agreed newspaper categories using labeled data from the news domain. The labeled news and the unlabeled blog corpus are highly dynamic and hourly growing with a topic drift, so a trade-off between accuracy and performance is required. Our approach is to apply a fast novel centroid-based algorithm, the Class-Feature-Centroid Classifier (CFC), to perform efficient cross-domain classification. Experiments showed that this algorithm achieves a comparable accuracy than k-NN …
Cited by:	2

:view less

Facet classification of blogs: Know-Center at the TREC 2009 blog distillation task

2009

Elisabeth Lex, Michael Granitzer, Andreas Juffinger
:read more

Authors:	Elisabeth Lex, Michael Granitzer, Andreas Juffinger
Publication date:	11/01/2009
Description:	In this paper, we outline our experiments carried out at the TREC 2009 Blog Distillation Task. Our system is based on a plain text index extracted from the XML feeds of the TREC Blogs08 dataset. This index was used to retrieve candidate blogs for the given topics. The resulting blogs were classified using a Support Vector Machine that was trained on a manually labelled subset of the TREC Blogs08 dataset. Our experiments included three runs on different features: firstly on nouns, secondly on stylometric properties, and thirdly on punctuation statistics. The facet identification based on our approach was successful, although a significant number of candidate blogs were not retrieved at all.
Cited by:	4
Resources:	[PDF] from researchgate.net

:view less

Efficient linear text segmentation based on information retrieval techniques

2009

Roman Kern, Michael Granitzer
Proceedings of the International Conference on Management of Emergent Digital EcoSystem…
:read more

Authors:	Roman Kern, Michael Granitzer
Publication date:	10/27/2009
Journal:	Proceedings of the International Conference on Management of Emergent Digital EcoSystems
Pages:	167-171
Description:	The task of linear text segmentation is to split a large text document into shorter fragments, usually blocks of consecutive sentences. The algorithms that demonstrated the best performance for this task come at the price of high computational complexity. In our work we present an algorithm that has a computational complexity of O(n) with n being the number of sentences in a document. The performance of our approach is evaluated against algorithms of higher complexity using standard benchmark data sets and we demonstrate that our approach provides comparable accuracy.
Cited by:	36
Resources:	[PDF] from acm.org

:view less

Evaluation of axiomatic approaches to crosslanguage retrieval

2009

Roman Kern, Andreas Juffinger, Michael Granitzer
Workshop of the Cross-Language Evaluation Forum for European Languages, 142-149
:read more

Authors:	Roman Kern, Andreas Juffinger, Michael Granitzer
Publication date:	09/30/2009
Journal:	Workshop of the Cross-Language Evaluation Forum for European Languages
Pages:	142-149
Publisher:	Springer Berlin Heidelberg
Description:	Integrating word sense disambiguation into an information retrieval system could potentially improve its performance. This is the major motivation for the Robust WSD tasks of the Ad-Hoc Track of the CLEF 2009 campaign. For these tasks we have build a customizable and flexible retrieval system. The best performing configuration of this system is based on research in the area of axiomatic approaches to information retrieval. Further, our experiments show that configurations that incorporate word sense disambiguation (WSD) information into the retrieval process did outperform those without. For the monolingual task the performance difference is more pronounced than for the bilingual task. Finally, we are able to show that our query translation approach does work effectively, even if applied in the monolingual task.
Cited by:	5
Resources:	[PDF] from cnr.it

:view less

Automatic cluster number selection using a split and merge k-means approach

2009

Markus Muhr, Michael Granitzer
2009 20th international workshop on database and expert systems application, 363-367
:read more

Authors:	Markus Muhr, Michael Granitzer
Publication date:	08/31/2009
Journal:	2009 20th international workshop on database and expert systems application
Pages:	363-367
Publisher:	IEEE
Description:	The k-means method is a simple and fast clustering technique that exhibits the problem of specifying the optimal number of clusters preliminarily. We address the problem of cluster number selection by using a k-means approach that exploits local changes of internal validity indices to split or merge clusters. Our split and merge k-means issues criterion functions to select clusters to be split or merged and fitness assessments on cluster structure changes. Experiments on standard test data sets show that this approach selects an accurate number of clusters with reasonable runtime and accuracy.
Cited by:	50
Resources:	[PDF] from semanticscholar.org

:view less

Visual knowledge discovery in dynamic enterprise text repositories

2009

Vedran Sabol, Wolfgang Kienreich, Markus Muhr, Werner Klieber, Michael Granitzer
2009 13th International Conference Information Visualisation, 361-368
:read more

Authors:	Vedran Sabol, Wolfgang Kienreich, Markus Muhr, Werner Klieber, Michael Granitzer
Publication date:	07/15/2009
Journal:	2009 13th International Conference Information Visualisation
Pages:	361-368
Publisher:	IEEE
Description:	Knowledge discovery involves data driven processes where data is transformed and processed by various algorithms to identify new knowledge. KnowMiner is a service oriented framework providing a rich set of knowledge discovery functionalities with focus on text data sets. Complementing results of automatic machine analysis with the immense processing power of human visual apparatus has the potential of significantly improving the process of acquiring new knowledge. VisTools is a lightweight visual analytics framework based on multiple coordinated views (MCV) paradigm designed for deployment atop the KnowMinerpsilas service architecture. In this paper we briefly present both frameworks and, driven by real-world customer requirements, describe how visual techniques can be synergistically combined with machine processing for effective analysis of dynamically changing, metadata-rich text documents …
Cited by:	22

:view less

ARS/SD: An Associative Retrieval Service for the Semantic Desktop

2009

Peter Scheir, Chiara Ghidini, Roman Kern, Michael Granitzer, Stefanie N Lindstaedt
Networked Knowledge-Networked Media: Integrating Knowledge Management, New Media Techno…
:read more

Authors:	Peter Scheir, Chiara Ghidini, Roman Kern, Michael Granitzer, Stefanie N Lindstaedt
Publication date:	07/10/2009
Journal:	Networked Knowledge-Networked Media: Integrating Knowledge Management, New Media Technologies and Semantic Systems
Pages:	95-111
Publisher:	Springer Berlin Heidelberg
Description:	While it is agreed that semantic enrichment of resources would lead to better search results, at present the low coverage of resources on the web with semantic information presents a major hurdle in realizing the vision of search on the Semantic Web. To address this problem we investigate how to improve retrieval performance in a setting where resources are sparsely annotated with semantic information. We suggest employing techniques from associative information retrieval to find relevant material, which was not originally annotated with the concepts used in a query. We present an associative retrieval service for the Semantic Desktop and evaluate if the use of associative retrieval techniques increases retrieval performance. Evaluation of new retrieval paradigms, as retrieval in the Semantic Web or on the Semantic Desktop, presents an additional challenge as no off-the-shelf test corpora for evaluation exist …

:view less

Accelerating k-means on the graphics processor via cuda

2009

Mario Zechner, Michael Granitzer
2009 First International Conference on Intensive Applications and Services, 7-15
:read more

Authors:	Mario Zechner, Michael Granitzer
Publication date:	04/20/2009
Journal:	2009 First International Conference on Intensive Applications and Services
Pages:	7-15
Publisher:	IEEE
Description:	In this paper an optimized k-means implementation on the graphics processing unit (GPU) is presented. NVIDIApsilas compute unified device architecture (CUDA), available from the G80 GPU family onwards, is used as the programming environment. Emphasis is placed on optimizations directly targeted at this architecture to best exploit the computational capabilities available. Additionally drawbacks and limitations of previous related work, e.g. maximum instance, dimension and centroid count are addressed. The algorithm is realized in a hybrid manner, parallelizing distance calculations on the GPU while sequentially updating cluster centroids on the CPU based on the results from the GPU calculations. An empirical performance study on synthetic data is given, demonstrating a maximum 14times speed increase to a fully SIMD optimized CPU implementation.
Cited by:	150
Resources:	[PDF] from researchgate.net

:view less

Blog credibility ranking by exploiting verified content

2009

Andreas Juffinger, Michael Granitzer, Elisabeth Lex
Proceedings of the 3rd workshop on Information credibility on the web, 51-58
:read more

Authors:	Andreas Juffinger, Michael Granitzer, Elisabeth Lex
Publication date:	04/20/2009
Journal:	Proceedings of the 3rd workshop on Information credibility on the web
Pages:	51-58
Description:	People use weblogs to express thoughts, present ideas and share knowledge. However, weblogs can also be misused to influence and manipulate the readers. Therefore the credibility of a blog has to be validated before the available information is used for analysis. The credibility of a blogentry is derived from the content, the credibility of the author or blog itself, respectively, and the external references or trackbacks. In this work we introduce an additional dimension to assess the credibility, namely the quantity structure. For our blog analysis system we derive the credibility therefore from two dimensions. Firstly, the quantity structure of a set of blogs and a reference corpus is compared and secondly, we analyse each separate blog content and examine the similarity with a verified news corpus. From the content similarity values we derive a ranking function. Our evaluation showed that one can sort out incredible …
Cited by:	131
Resources:	[PDF] from acm.org

:view less

Distributed Web 2.0 crawling for ontology evolution.

2009

Andreas Juffinger, Thomas Neidhart, Michael Granitzer, Roman Kern, Albert Weichselbraun…
Journal of Digital Information Management 7 (2), 114-120
:read more

Authors:	Andreas Juffinger, Thomas Neidhart, Michael Granitzer, Roman Kern, Albert Weichselbraun, Gerhard Wohlgenannt, Arno Scharl
Publication date:	04/01/2009
Journal:	Journal of Digital Information Management
Volume:	7
Issue:	2
Pages:	114-120
Publisher:	Digital Information Research Foundation
Description:	The World Wide Web as a social network reflects changes of interest in certain domains. It has been shown that free online content available through blogs, wikis, news media and online forums is a valuable source of information to identify trends in certain domains. Utilizing this data, one can construct ontologies that describe this information and provide a semantically correct overview of a domain. Tracked over time this also enables a user to identify trends and hypes. The decentralised structure of the Internet, the huge amount of data and upcoming Web2. 0 technologies pose several challenges to a crawling system for ontology learning, evolution and trend analysis. This paper presents a distributed crawling system with browser integration for Web2. 0. The proposed crawler is a high performanceWeb data retrieval system aimed to gather browser-equivalent textual Web content and prepare it for ontology learning.
Cited by:	1

:view less

Knowledge discovery using the KnowMiner framework

2009

Werner Klieber, Vedran Sabol, Markus Muhr, Roman Kern, Georg Öttl, Michael Granitzer
Proc. IADIS 9
:read more

Authors:	Werner Klieber, Vedran Sabol, Markus Muhr, Roman Kern, Georg Öttl, Michael Granitzer
Publication date:	02/01/2009
Journal:	Proc. IADIS
Volume:	9
Description:	Services for increasing information quality in unstructured information are central to mostly all applications of information management. Service-oriented computing is a promising paradigm for embedding such intelligent services in different application scenarios and heterogeneous IT landscapes. In this paper we present our knowledge discovery framework named “KnowMiner”, a service oriented architecture designed with the primary goal to ease usage for non knowledge discovery experts while providing a rich set of knowledge discovery functionalities for very different application scenarios. We will introduce the underlying data model of the framework, its basic services as well as concepts for workflow composition based on pre-and post-conditions. We summarize our findings on the development and application of such a framework in a set of design aspects and outline that formal pre-and post-conditions are not suited for data intensive applications with a high spectrum of functionality.
Cited by:	24
Resources:	[PDF] from researchgate.net

:view less

Application of Axiomatic Approaches to Crosslanguage Retrieval.

2009

Roman Kern, Andreas Juffinger, Michael Granitzer
CLEF 2009, 0-0
:read more

Authors:	Roman Kern, Andreas Juffinger, Michael Granitzer
Publication date:	01/01/2009
Journal:	CLEF 2009
Pages:	0-0
Publisher:	.
Description:	Application of Axiomatic Approaches to Crosslanguage Retrieval. - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Application of Axiomatic Approaches to Crosslanguage Retrieval. Roman Kern, Andreas Juffinger, Michael Granitzer Know Center Research GmbH (98770) Publikation: Beitrag in Buch/Bericht/Konferenzband › Beitrag in einem Konferenzband › Begutachtung Übersicht Originalsprache englisch Titel CLEF 2009 Herausgeber (Verlag) . Seiten 0-0 Publikationsstatus Veröffentlicht - 2009 Dieses zitieren APA Standard Harvard Vancouver Author BIBTEX RIS Angetrieben von Pure, Scopus & Elsevier Fingerprint Engine™ …
Cited by:	1

:view less

Wissensmanagement mit DYONIPOS: Konzepte, Algorithmen und Technologien

2009

Michael Granitzer, Josef Makolm, Doris Ipsmiller
Internationales Rechtsinformatik Symposion (IRIS 09)(2009), 1-5
:read more

Authors:	Michael Granitzer, Josef Makolm, Doris Ipsmiller
Publication date:	01/01/2009
Journal:	Internationales Rechtsinformatik Symposion (IRIS 09)(2009)
Pages:	1-5
Publisher:	.
Description:	Wissensmanagement mit DYONIPOS: Konzepte, Algorithmen und Technologien - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Wissensmanagement mit DYONIPOS: Konzepte, Algorithmen und Technologien Michael Granitzer, Josef Makolm, Doris Ipsmiller Publikation: Beitrag in Buch/Bericht/Konferenzband › Beitrag in einem Konferenzband › Begutachtung Übersicht Originalsprache deutsch Titel Internationales Rechtsinformatik Symposion (IRIS 09) (2009) Herausgeber (Verlag) . Seiten 1-5 Publikationsstatus Veröffentlicht - 2009 Veranstaltung Internationales Rechtsinformatik Symposion (IRIS 09) (2009) - Salzburg, …

:view less

Spectral web content trend analysis

2009

Andreas Juffinger, Elisabeth Lex, Reinhard Willfort, Michael Granitzer
IADIS International Conference on WWW/Internet, 0-0
:read more

Authors:	Andreas Juffinger, Elisabeth Lex, Reinhard Willfort, Michael Granitzer
Publication date:	01/01/2009
Journal:	IADIS International Conference on WWW/Internet
Pages:	0-0
Publisher:	.
Description:	Spectral web content trend analysis - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Search content at Graz University of Technology Home Persons Publications Organisations Projects Activities Prizes Press/Media Spectral web content trend analysis Andreas Juffinger , Elisabeth Lex , Reinhard Willfort , Michael Granitzer Know Center Research GmbH (98770) Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review Overview Original language English Title of host publication Proceedings of IADIS International Conference WWWInternet (2009) Publisher . Pages 0-0 Publication status Published - 2009 Event IADIS International Conference on WWW/Internet - Rom, Italy Duration: 19 Nov 2009 → 22 Nov 2009 Conference Conference IADIS …
Cited by:	1

:view less

A Wikipedia-Based Multilingual Retrieval Model

2009

RK Andreas Juffinger, M Granitzer
Evaluating Systems for Multilingual and Multimodal Information Access, 155-162
:read more

Authors:	RK Andreas Juffinger, M Granitzer
Publication date:	01/01/2009
Journal:	Evaluating Systems for Multilingual and Multimodal Information Access
Pages:	155-162
Cited by:	5

:view less

Using ontologies for software documentation

2009

Werner Klieber, Vedran Sabol, Roman Kern, Markus Muhr, Michael Granitzer
Malaysian Joint Conference on Artificial Intelligence-MJCAI2009
:read more

Authors:	Werner Klieber, Vedran Sabol, Roman Kern, Markus Muhr, Michael Granitzer
Publication date:	01/01/2009
Journal:	Malaysian Joint Conference on Artificial Intelligence-MJCAI2009
Description:	High quality software documentation is a substantial issue to understand software systems. Shorter time-to-market software cycles increase the importance of automatism to keep the documentation up to date. In this paper, we describe the automatic support of the software documentation process using a social semantic software approach. Therefore, we introduce a software documentation ontology as underlying knowledge base. The defined ontology is populated automatically through analyzing source code, documentation and code executions and made accessible to developers via a wiki. The wiki enforces collaboration and allows correcting analysis errors. We demonstrate, that the use of social semantic systems can support software documentation processes efficiently.
Cited by:	7
Resources:	[PDF] from researchgate.net

:view less

Automated Blog Classification: A Cross-Domain Approach

2009

Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
IADIS International Conference WWW/Internet 2009, 598-605
:read more

Authors:	Elisabeth Lex, Christin Seifert, Michael Granitzer, Andreas Juffinger
Publication date:	01/01/2009
Journal:	IADIS International Conference WWW/Internet 2009
Pages:	598-605
Publisher:	IADIS
Description:	The automated classification of blogs is highly important for the relatively new field of blog analysis. To classify blogs into topics or other categories, usually supervised text classification algorithms are applied. However, supervised text classifiers need a sufficient large amount of labeled data to learn a good model. Especially for blogs, data labeled with terms that capture current and actual topics are not available and data labeled in the past is usually not applicable due to topic drifts. Besides, tagged blogs collected from the web exhibit a vocabulary that is rather heterogeneous, diverse and not commonly agreed upon. In our work, we focus on news-related blogs dealing with current events. Our goal is to classify blog posts into given, common newspaper categories. As a baseline, we have high quality labeled data from a German news corpus. Our approach is to exploit the labeled data from the news corpus and use this knowledge to perform crossdomain classification on the unlabeled blogs. We need a solution with high performance, because both our corpora are dynamic and our classifier model needs to be up-to-date. In this work, we evaluated a number of text classification algorithms with different parameter settings by means of accuracy and complexity. Qualitative and quantitative analysis revealed that a recently proposed centroid-based algorithm, the Class-Feature-Centroid classifier (CFC), serves best for our setting because it achieves a comparable accuracy with state-of-the-art text classifiers and outperforms all other algorithms regarding complexity and memory consumption.
Cited by:	5
Resources:	[PDF] from researchgate.net

:view less

K-means on the graphics processor: design and experimental analysis

2009

Mario Zechner, Michael Granitzer
Int. J. Adv. Syst. Meas 2, 224-235
:read more

Authors:	Mario Zechner, Michael Granitzer
Publication date:	01/01/2009
Journal:	Int. J. Adv. Syst. Meas
Volume:	2
Pages:	224-235
Description:	Apart from algorithmic improvements many intensive machine learning algorithms can gain performance by parallelization. Programmable graphics processing units (GPU) offer a highly data parallel architecture that is suitable for many computational tasks in machine learning. We present an optimized k-means implementation on the graphics processing unit. NVIDIA’s Compute Unified Device Architecture (CUDA), available from the G80 GPU family onwards, is used as the programming environment. Emphasis is placed on optimizations directly targeted at this architecture to best exploit the computational capabilities available. Additionally drawbacks and limitations of previous related work, eg maximum instance, dimension and centroid count are addressed. The algorithm is realized in a hybrid manner, parallelizing distance calculations on the GPU while sequentially updating cluster centroids on the CPU based on the results from the GPU calculations. An empirical performance study on synthetic data is given, demonstrating a maximum 14x speed increase to a fully SIMD optimized CPU implementation. We present detailed empirical data on the runtime behavior of the various stages of the implementation, identify bottlenecks and investigate potential discrepancies arising from different rounding modes on the GPU and CPU based. We extend our previous work in [1] by giving a more in depth description of CUDA as well as including previously omitted experimental data.
Cited by:	6
Resources:	[PDF] from researchgate.net

:view less

Know-Center at TREC 2009 Blog Distillation Task: A Notebook Paper

2009

Elisabeth Lex, Michael Granitzer, Andreas Juffinger
:read more

Authors:	Elisabeth Lex, Michael Granitzer, Andreas Juffinger
Publication date:	01/01/2009
Description:	This paper outlines our experiments carried out at TREC 2009 Blog Distillation Task. Our system is based on a plain text index extracted from the XML feeds of the TREC Blogs08 dataset. This index was used to retrieve candidate blogs for the given topics. The resulting blogs were classified using a Support Vector Machine that was trained on a manually labelled subset of the TREC Blogs08 dataset. Three runs were conducted, one based on nouns, one based on stylometric properties, and one based on punctuation statistics. The facet identification based on our approach was successful, although a significant number of candidate blogs were not retrieved at all.
Resources:	[PDF] from researchgate.net

:view less

Machine learning based work task classification

2009

Michael Granitzer, Andreas S Rath, Mark Kröll, Christin Seifert, Doris Ipsmiller, Didie…
Journal of Digital Information Management 7 (5), 306-313
:read more

Authors:	Michael Granitzer, Andreas S Rath, Mark Kröll, Christin Seifert, Doris Ipsmiller, Didier Devaurs, Nicolas Weber, Stefanie Lindstaedt
Publication date:	01/01/2009
Journal:	Journal of Digital Information Management
Volume:	7
Issue:	5
Pages:	306-313
Description:	Increasing the productivity of a knowledge worker via intelligent applications requires the identification of a user's current work task, i.e. the current work context a user resides in. In this work we present and evaluate machine learning based work task detection methods. By viewing a work task as sequence of digital interaction patterns of mouse clicks and key strokes, we present (i) a methodology for recording those user interactions and (ii) an in-depth analysis of supervised classification models for classifying work tasks in two different scenarios: a task centric scenario and a user centric scenario. We analyze different supervised classification models, feature types and feature selection methods on a laboratory as well as a real world data set. Results show satisfiable accuracy and high user acceptance by using relatively simple types of features.
Cited by:	19
Resources:	[PDF] from hal.science

:view less

Taxonomy extraction from german encyclopedic texts

2009

Michael Granitzer, Andreas Augustin, Wolfgang Kienreich, Vedran Sabol
Proceedings of the Malaysian Joint Conference on Artificial Intelligence
:read more

Authors:	Michael Granitzer, Andreas Augustin, Wolfgang Kienreich, Vedran Sabol
Publication date:	01/01/2009
Journal:	Proceedings of the Malaysian Joint Conference on Artificial Intelligence
Description:	This paper presents a procedure to construct robust taxonomies from natural language German encyclopedic text. Taxonomic relations are extracted through Hearst patterns, validated via search engines and incorporated into an base ontology. By verifying the relations using an external, web-based source of evidence, the procedure extracted an accurate taxonomy with minimal human intervention. Experimental results revealed that high precision values can be achieved. Results show that a well grounded suggestion of solid patterns for the extraction of German hypernyms relations is possible and that the chosen approach may be efficiently transfered to other western languages.
Cited by:	6
Resources:	[PDF] from researchgate.net

:view less

Discovery and evaluation of non-taxonomic relations in domain ontologies

2009

Albert Weichselbraun, Gerhard Wohlgenannt, Arno Scharl, Michael Granitzer, Thomas Neidh…
International Journal of Metadata, Semantics and Ontologies 4 (3), 212-222
:read more

Authors:	Albert Weichselbraun, Gerhard Wohlgenannt, Arno Scharl, Michael Granitzer, Thomas Neidhart, Andreas Juffinger
Publication date:	01/01/2009
Journal:	International Journal of Metadata, Semantics and Ontologies
Volume:	4
Issue:	3
Pages:	212-222
Publisher:	Inderscience Publishers
Description:	The identification and labelling of non-hierarchical relations are among the most challenging tasks in ontology learning. This paper describes a bottom-up approach for automatically suggesting ontology link types. The presented method extracts verb-vectors from semantic relations identified in the domain corpus, aggregates them by computing centroids for known relation types, and stores the centroids in a central knowledge base. Comparing verb-vectors extracted from unknown relations with the stored centroids yields link type suggestions. Domain experts evaluate these suggestions, refining the knowledge base and constantly improving the component's accuracy. A final evaluation provides a detailed statistical analysis of the introduced approach.
Cited by:	37
Resources:	[PDF] from weblyzard.com

:view less

Context Resolution Strategies for Automatic Wikipedia Linking

2008

Michael Granitzer, Christin Seifert, Mario Zechner
INEX 2008 pre-proceedings, 0-0
:read more

Authors:	Michael Granitzer, Christin Seifert, Mario Zechner
Publication date:	12/15/2008
Journal:	INEX 2008 pre-proceedings
Pages:	0-0
Publisher:	.
Description:	Automatically linking Wikipedia pages is done mostly by two strategies:(i) a content based strategy based on word similarities or (ii) a structural similarity exploiting link characteristics. In our approach we focus on a content based strategy by finding anchors using the title of candidate Wikipedia pages and resolving matching links by taking the context of the link anchor, ie its surrounding text, into account. Bestentry-points are estimated on a combination of title and content based similarity. Our goal was to evaluate syntactic title matching properties and the influence of the context around anchors for disambiguation and best-entry-point detection. Results show, that the whole Wikipedia page provides the best context for resolving links and that simple inverse document frequency based scoring of anchor texts is also capable of achieving high accuracy.
Cited by:	1
Resources:	[PDF] from utwente.nl

:view less

Context based wikipedia linking

2008

Michael Granitzer, Christin Seifert, Mario Zechner
International Workshop of the Initiative for the Evaluation of XML Retrieval, 354-365
:read more

Authors:	Michael Granitzer, Christin Seifert, Mario Zechner
Publication date:	12/15/2008
Journal:	International Workshop of the Initiative for the Evaluation of XML Retrieval
Pages:	354-365
Publisher:	Springer Berlin Heidelberg
Description:	Automatically linking Wikipedia pages can be done either content based by exploiting word similarities or structure based by exploiting characteristics of the link graph. Our approach focuses on a content based strategy by detecting Wikipedia titles as link candidates and selecting the most relevant ones as links. The relevance calculation is based on the context, i.e. the surrounding text of a link candidate. Our goal was to evaluate the influence of the link-context on selecting relevant links and determining a links best-entry-point. Results show, that a whole Wikipedia page provides the best context for resolving link and that straight forward inverse document frequency based scoring of anchor texts achieves around 4% less Mean Average Precision on the provided data set.
Cited by:	11
Resources:	[PDF] from researchgate.net

:view less

Context Resolution Strategies for Automatic Wikipedia Learning

2008

Michael Granitzer, Christin Seifert, Mario Zechner
7th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX …
:read more

Authors:	Michael Granitzer, Christin Seifert, Mario Zechner
Publication date:	12/01/2008
Journal:	7th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2008
Description:	Automatically linking Wikipedia pages is done mostly by two strategies:(i) a content based strategy based on word similarities or (ii) a structural similarity exploiting link characteristics. In our approach we focus on a content based strategy by finding anchors using the title of candidate Wikipedia pages and resolving matching links by taking the context of the link anchor, ie its surrounding text, into account. Bestentry-points are estimated on a combination of title and content based similarity. Our goal was to evaluate syntactic title matching properties and the influence of the context around anchors for disambiguation and best-entry-point detection. Results show, that the whole Wikipedia page provides the best context for resolving links and that simple inverse document frequency based scoring of anchor texts is also capable of achieving high accuracy.
Resources:	[PDF] from utwente.nl

:view less

A generic framework for visualizing the news article domain and its application to real-world data

2008

E Lex, C Seifert, W Kienreich, M Granitzer
Journal of Digital Information Management 6 (6), 434-442
:read more

Authors:	E Lex, C Seifert, W Kienreich, M Granitzer
Publication date:	12/01/2008
Journal:	Journal of Digital Information Management
Volume:	6
Issue:	6
Pages:	434-442
Description:	In this work we present APA Labs, a generic framework for visualizing the news article domain. APA Labs is a web-based platform enabling retrieval and analysis of news repositories provided by the Austrian Press Agency. APA Labs is designed as a rich internet application combined with a modular system of interactive visualizations. News articles are analyzed using domain specific named entity extraction methods combined with language specific heuristics. The proposed methods were subject to an evaluation procedure outlined in this contribution. This article illustrates the domain, the underlying concepts and implementation details. Several visualization modules are presented and an outlook on planned modules is given. Being online for around six months the community feedback as well as the easy integration of new modules shows the success of the underlying concept and the platform itself.
Cited by:	8
Resources:	[PDF] from researchgate.net

:view less

Analysis of machine learning techniques for context extraction

2008

Michael Granitzer, Mark Kroll, Christin Seifert, Andreas S Rath, Nicolas Weber, Olivia …
2008 Third International Conference on Digital Information Management, 233-240
:read more

Authors:	Michael Granitzer, Mark Kroll, Christin Seifert, Andreas S Rath, Nicolas Weber, Olivia Dietzel, Stefanie Lindstaedt
Publication date:	11/13/2008
Journal:	2008 Third International Conference on Digital Information Management
Pages:	233-240
Publisher:	IEEE
Description:	‘Context is key’ conveys the importance of capturing the digital environment of a knowledge worker. Knowing the user’s context offers various possibilities for support, like for example enhancing information delivery or providing work guidance. Hence, user interactions have to be aggregated and mapped to predefined task categories. Without machine learning tools, such an assignment has to be done manually. The identification of suitable machine learning algorithms is necessary in order to ensure accurate and timely classification of the user’s context without inducing additional workload. This paper provides a methodology for recording user interactions and an analysis of supervised classification models, feature types and feature selection for automatically detecting the current task and context of a user. Our analysis is based on a real world data set and shows the applicability of machine learning techniques.
Cited by:	25
Resources:	[PDF] from researchgate.net

:view less

Context-Aware Knowledge Services CHI 2008 Workshop on Personal Information Management

2008

A Rath, N Weber, M Kröll, M Granitzer, O Dietzel, S Lindstaedt
Florence, Italy http://pim2008. ethz. ch/papers/pim2008-rath-etal. pdf accessed 26
:read more

Authors:	A Rath, N Weber, M Kröll, M Granitzer, O Dietzel, S Lindstaedt
Publication date:	11/01/2008
Journal:	Florence, Italy http://pim2008. ethz. ch/papers/pim2008-rath-etal. pdf accessed
Volume:	26
Cited by:	2

:view less

Evaluation of Automatic Linking Strategies for Wikipedia Pages

2008

Michael Granitzer, Mario Zechner, Christin Seifert, Josef Kolbitsch, Peter Kemper
IADIS International Conference WWW/Internet 2008, 243-250
:read more

Authors:	Michael Granitzer, Mario Zechner, Christin Seifert, Josef Kolbitsch, Peter Kemper
Publication date:	10/01/2008
Journal:	IADIS International Conference WWW/Internet 2008
Pages:	243-250
Publisher:	IADIS
Description:	Wikipedia contains an enormous amount of human knowledge. The wide range of covered topics is hierarchically organized in categories and strongly inter-linked. Its structure, its size and the fact that it is generated by humans are the reasons for the attention Wikipedia receives from researchers in different fields. Especially the link structure of Wikipedia is of huge importance not only for humans browsing the collection, but also as a resource for bootstrapping machine intelligence and the semantic web. Motivated by the fact that manual maintenance and creation of hyperlinks is labor intensive, this paper explores properties for automatic link creation between Wikipedia pages in this paper. Focusing on ad-hoc linking approaches we evaluate linking strategies on the word as well as on the document level using a standard test data set. As it is shown, rather simple approaches yield to reliable results and may be applicable in different application scenarios. Disambiguation strategies based on standard IR techniques help to boost accuracy delivering reasonable results.
Resources:	[PDF] from utwente.nl

:view less

TU Graz: Course: 707.000 Web Science and Web Technology: Lecture 10: Text Mining

2008

Michael Granitzer
:read more

Authors:	Michael Granitzer
Publication date:	09/30/2008
Description:	This class introduces basics of web mining and information retrieval including, for example, an introduction to the Vector Space Model and Text Mining. Guest Lecturer: Dr. Michael Granitzer Optional: Modeling the Internet and the Web: Probabilistic Methods and Algorithms, Pierre Baldi, Paolo Frasconi, Padhraic Smyth, Wiley, 2003 (Chapter 4, Text Analysis)
Resources:	[PDF] from soton.ac.uk

:view less

Crosslanguage retrieval based on wikipedia statistics

2008

Andreas Juffinger, Roman Kern, Michael Granitzer
Workshop of the Cross-Language Evaluation Forum for European Languages, 155-162
:read more

Authors:	Andreas Juffinger, Roman Kern, Michael Granitzer
Publication date:	09/17/2008
Journal:	Workshop of the Cross-Language Evaluation Forum for European Languages
Pages:	155-162
Publisher:	Springer Berlin Heidelberg
Description:	In this paper we present the methodology, implementations and evaluation results of the crosslanguage retrieval system we have developed for the Robust WSD Task at CLEF 2008. Our system is based on query preprocessing for translation and homogenisation of queries. The presented preprocessing of queries includes two stages: Firstly, a query translation step based on term statistics of cooccuring articles in Wikipedia. Secondly, different disjunct query composition techniques to search in the CLEF corpus. We apply the same preprocessing steps for the monolingual as well as the crosslingual task and thereby acting fair and in a similar way across these tasks. The evaluation revealed that the similar processing comes at nearly no costs for monolingual retrieval but enables us to do crosslanguage retrieval and also a feasible comparison of our system performance on these two tasks.
Cited by:	5

:view less

Exploiting Cooccurrence on Corpus and Document Level for Fair Crosslanguage Retrieval.

2008

Andreas Juffinger, Roman Kern, Michael Granitzer, Graz Know-Center
CLEF (Working Notes)
:read more

Authors:	Andreas Juffinger, Roman Kern, Michael Granitzer, Graz Know-Center
Publication date:	09/17/2008
Journal:	CLEF (Working Notes)
Description:	In this paper we describe the methodology, architecture and implementation of the information retrieval system we have developed for the Robust WSD Task at CLEF 2008. Our system is based on an extensive query preprocessing step for homogenisation of the corpus queries. The preprocessing of queries includes: firstly, an query expansion step based on Wordnet Synonsyms or an Associative Index, secondly a query translation step based on corpus article cooccurrence in Wikipedia, and thirdly a standard disjunct index search in the CLEF corpus. The crosslanguage enabled system behaves thereby as much as possible fair over different languages. We apply the same preprocessing steps, independent of the query and corpus language, to all queries.
Cited by:	4
Resources:	[PDF] from cnr.it

:view less

Automating knowledge transfer and creation in knowledge intensive business processes

2008

Michael Granitzer, Gisela Granitzer, Klaus Tochtermann, Stefanie Lindstaedt, Andreas Ra…
International Conference on Business Process Management, 678-686
:read more

Authors:	Michael Granitzer, Gisela Granitzer, Klaus Tochtermann, Stefanie Lindstaedt, Andreas Rath, Wolfgang Groiß
Publication date:	09/01/2008
Journal:	International Conference on Business Process Management
Pages:	678-686
Publisher:	Springer Berlin Heidelberg
Description:	It is a well known fact that a wealth of knowledge lies in the head of employees making them one of the most or even the most valuable asset of organisations. But often this knowledge is not documented and organised in knowledge systems as required by the organisation, but informally shared. Of course this is against the organisation’s aim for keeping knowledge reusable as well as easily and permanently available independent of individual knowledge workers. In this contribution we suggest a solution which captures the collective knowledge to the benefit of the organisation and the knowledge worker. By automatically identifying activity patterns and aggregating them to tasks as well as by assigning resources to these tasks, our proposed solution fulfils the organisation’s need for documentation and structuring of knowledge work. On the other hand it fulfils the the knowledge worker’s need for relevant, currently …
Cited by:	7
Resources:	[PDF] from researchgate.net

:view less

On the beauty and usability of tag clouds

2008

Christin Seifert, Barbara Kump, Wolfgang Kienreich, Gisela Granitzer, Michael Granitzer
2008 12th International Conference Information Visualisation, 17-25
:read more

Authors:	Christin Seifert, Barbara Kump, Wolfgang Kienreich, Gisela Granitzer, Michael Granitzer
Publication date:	07/09/2008
Journal:	2008 12th International Conference Information Visualisation
Pages:	17-25
Publisher:	IEEE
Description:	Tag clouds are text-based visual representations of a set of tags usually depicting tag importance by font size. Recent trends in social and collaborative software have greatly increased the popularity of this type of visualization. This paper proposes a family of novel algorithms for tag cloud layout and presents evaluation results obtained from an extensive user study and a technical evaluation. The algorithms address issues found in many common approaches, for example large whitespaces, overlapping tags and restriction to specific boundaries. The layouts computed by these algorithms are compact and clear, have small whitespaces and may feature arbitrary convex polygons as boundaries. The results of the user study and the technical evaluation enable designers to devise a combination of algorithm and parameters which produces satisfying tag cloud layouts for many application scenarios.
Cited by:	176

:view less

Multimedia semantics-the role of metadata

2008

Michael Granitzer, Mathias Lux, Marc Spaniol
:read more

Authors:	Michael Granitzer, Mathias Lux, Marc Spaniol
Publication date:	05/27/2008
Publisher:	Springer
Description:	Research in the field of multimedia metadata is especially challenging: Lots of scientific publications and reports on research projects are published every year and the range of possible applications is diverse and huge. This book gives an overview on fundamental issues within the field of multimedia metadata focusing on contextualized, ubiquitous, accessible and interoperable services on a higher semantic level. The book in hand provides a selection of basic articles being a base for multimedia metadata research. Furthermore it presents a view on the current state of the art in multimedia metadata research. It provides information from versatile applications domains (Broadcasting, Interactive TV, E-Learning and Social Software) such as: Multimedia on the Web 2.0-Databases for Multimedia (Meta-) Data-Multimedia Information Retrieval and Evaluation-Multimedia Metadata Standards-Ontologies for Multimedia. The multimedia metadata community (www. multimedia-metadata. info), wherefrom this book originated, brings together experts from research and industry in the area of multimedia metadata research and application development. The community bridges the gap between an academic research and an industrial scale development of innovative products. By summarizing the work of the community this book contributes to the aforementioned fields by addressing these topics for a broad range of readers.
Cited by:	9

:view less

Visualisation techniques for analysis and exploration of multimedia data

2008

Vedran Sabol, Wolfgang Kienreich, Michael Granitzer
Multimedia Semantics—The Role of Metadata, 219-238
:read more

Authors:	Vedran Sabol, Wolfgang Kienreich, Michael Granitzer
Publication date:	05/27/2008
Journal:	Multimedia Semantics—The Role of Metadata
Pages:	219-238
Publisher:	Springer Berlin Heidelberg
Description:	Recent studies estimate the amount of information globally available to grow from 5 Exabyte in 2003 [29] and 161 Exabyte in 2006 to 998 Exabyte until 2010 [17]. Most of the new information created, shared and archived is of a multimedia nature. For example, 33,000 worldwide TV stations produced approximately 48 million hours of broadcasted video content in 2002 [51]. This already impressive number acquires still more significance considering that a mere 2000 hours of CNN news videos consist of approximately 1.7 million shots and almost 70,000 stories [52]. Another, less obvious, example of multimedia information influx is provided by video surveillance at some 14,000 worldwide air terminals, which produces 4.5 million hours of video per day [51]. To quantify the distribution of information types, 75% of available content consisted of images and video streams and 20% was contributed by voice recordings in …
Cited by:	8

:view less

Extending folksonomies for image tagging

2008

Roman Kern, Michael Granitzer, Viktoria Pammer
2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services…
:read more

Authors:	Roman Kern, Michael Granitzer, Viktoria Pammer
Publication date:	05/07/2008
Journal:	2008 Ninth International Workshop on Image Analysis for Multimedia Interactive Services
Pages:	126-129
Publisher:	IEEE
Description:	Due to the unsatisfactory results of content based image retrieval methods, organisation and retrieval of multimedia data strongly relies on metadata and free text description. Folksonomies, collaboratively created sets of metadata, merged recently and help organising multimedia information on the Web. Our contribution addresses the question how to extend a classical folksonomy with additional metadata. We also discuss the quality of the extended folksonomy and its application for tag recommendation. We show that some relations of the original folksonomies can be replaced while others are unique. In addition our analysis shows, that for 40% the correct tag is in the first 10% of the set of tag recommendations.
Cited by:	17

:view less

Text mapping: Visualising unstructured, structured, and time-based text collections

2008

Vedran Sabol, Keith Andrews, Wolfgang Kienreich, Michael Granitzer
Intelligent Decision Technologies 2 (2), 117-128
:read more

Authors:	Vedran Sabol, Keith Andrews, Wolfgang Kienreich, Michael Granitzer
Publication date:	05/01/2008
Journal:	Intelligent Decision Technologies
Volume:	2
Issue:	2
Pages:	117-128
Publisher:	SAGE Publications
Description:	Large collections of text documents are increasingly common, both in business and personal information environments. Tools from the field of information visualisation are being used to help users make sense of and extract useful knowledge from such collections. Flat text collections are often visualised using distance calculations between documents and subsequent (distance-preserving) projection. Distance calculations are often based on a vector space of term vectors. Projection is often achieved with a force-directed placement algorithm. Where extra information about a text collection is available, such as a topical hierarchy or some chronological ordering, it can be used to improve a visualisation. This paper gives an overview of text mapping techniques.
Cited by:	1
Resources:	[PDF] from archive.org

:view less

of Book: Encyclopedia of Multimedia

2008

Werner Bailer, Lionel Brunie, Mario Döller, Michael Granitzer, Ralf Klamma, Harald Kosc…
:read more

Authors:	Werner Bailer, Lionel Brunie, Mario Döller, Michael Granitzer, Ralf Klamma, Harald Kosch, Mathias Lux, Marc Spaniol
Publication date:	01/01/2008
Publisher:	Springer

:view less

KnowMiner: Konzeption & Entwicklung eines generischen Wissenserschließungsframeworks

2008

Michael Granitzer
:read more

Authors:	Michael Granitzer
Publication date:	01/01/2008
Publisher:	VDM Verlag Dr. Müller
Description:	KnowMiner: Konzeption & Entwicklung eines generischen Wissenserschließungsframeworks - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Home Persons Publications Organisations Projects Activities Prizes Press/Media Search by expertise, name or affiliation KnowMiner: Konzeption & Entwicklung eines generischen Wissenserschließungsframeworks Michael Granitzer Research output: Book/Report › Book Overview Original language German Place of Publication Saarbrücken Publisher VDM Verlag Dr. Müller Edition 1 ISBN (Print) 978-3-836-47209-8 Publication status Published - 2008 Cite this APA Standard Harvard Vancouver Author BIBTEX RIS Powered by Pure, Scopus & Elsevier Fingerprint Engine™ All content on this site: Copyright © 2025 Graz University of Technology, …
Cited by:	9

:view less

Wer gesucht wird, wird gefunden!-Web 2.0 und der Umgang mit sensiblen Daten

2008

Gisela Granitzer, Klaus Tochtermann, Michael Granitzer
Watching you-Kontrolle, Datenhandel, Überwachung, 56-62
:read more

Authors:	Gisela Granitzer, Klaus Tochtermann, Michael Granitzer
Publication date:	01/01/2008
Journal:	Watching you-Kontrolle, Datenhandel, Überwachung
Pages:	56-62
Publisher:	Leykam Buchverlagsgesellschaft
Description:	Wer gesucht wird, wird gefunden! - Web 2.0 und der Umgang mit sensiblen Daten - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Home Persons Publications Organisations Projects Activities Prizes Press/Media Search by expertise, name or affiliation Wer gesucht wird, wird gefunden! - Web 2.0 und der Umgang mit sensiblen Daten Gisela Granitzer, Klaus Tochtermann, Michael Granitzer Know Center Research GmbH (98770) Research output: Chapter in Book/Report/Conference proceeding › Chapter Overview Original language German Title of host publication Watching you - Kontrolle, Datenhandel, Überwachung Place of Publication Graz Publisher Leykam Buchverlagsgesellschaft Pages 56-62 Volume 11 Edition 1 ISBN (Print) 978-3-7011-7638-0 Publication status Published - …

:view less

Die clevere Art der Suche

2008

Gisela Granitzer, Klaus Tochtermann, Michael Granitzer
E-Commerce-Magazin (7), 0-0
:read more

Authors:	Gisela Granitzer, Klaus Tochtermann, Michael Granitzer
Publication date:	01/01/2008
Journal:	E-Commerce-Magazin
Issue:	7
Pages:	0-0
Description:	Die clevere Art der Suche - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Search content at Technische Universität Graz Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Die clevere Art der Suche Gisela Granitzer, Klaus Tochtermann, Michael Granitzer Publikation: Beitrag in einer Fachzeitschrift › Artikel Übersicht Originalsprache deutsch Seiten (von - bis) 0-0 Fachzeitschrift E-Commerce-Magazin Ausgabenummer 7 Publikationsstatus Veröffentlicht - 2008 Dieses zitieren APA Standard Harvard Vancouver Author BIBTEX RIS Angetrieben von Pure, Scopus & Elsevier Fingerprint Engine™ Alle Inhalte auf dieser Website: Copyright © 2026 Technische Universität Graz, zugehörige Lizenzgebende und Mitwirkende. Alle Rechte …

:view less

Multimedia metadata standards

2008

Werner Bailer, Lionel Brunie, Mario Döller, Michael Granitzer, Ralf Klamma, Harald Kosc…
Encyclopedia of Multimedia, 568-575
:read more

Authors:	Werner Bailer, Lionel Brunie, Mario Döller, Michael Granitzer, Ralf Klamma, Harald Kosch, Mathias Lux, Marc Spaniol
Publication date:	01/01/2008
Journal:	Encyclopedia of Multimedia
Pages:	568-575
Publisher:	Springer Verlag
Description:	Multimedia metadata describes various aspects of multimedia content, including formal and technical properties (e.g. encoding, format), information about the creation of content, the processing applied, its use, rights information and the structure and semantics of the content itself. The importance of these aspects depends on the role of the provider or consumer of the metadata and the stage in the media life cycle. Multimedia metadata is among others important for production, distribution, search and indexing of multimedia content.
Cited by:	7
Resources:	[PDF] from hal.science

:view less

Context-aware knowledge services

2008

Andreas S Rath, Nicolas Weber, Mark Kröll, Michael Granitzer, Olivia Dietzel, Stefanie …
Workshop on personal information management, chi 8
:read more

Authors:	Andreas S Rath, Nicolas Weber, Mark Kröll, Michael Granitzer, Olivia Dietzel, Stefanie N Lindstaedt
Publication date:	01/01/2008
Journal:	Workshop on personal information management, chi
Volume:	8
Description:	Improving the productivity of knowledge workers is an open research challenge. Our approach is based on providing a large variety of knowledge services which take the current work task and information need (work context) of the knowledge worker into account. In the following we present the DYONIPOS application which strives to automatically identify a user’s work task and then contextualizes different types of knowledge services accordingly. These knowledge services then provide information (documents, people, locations) both from the user’s personal as well as from the organizational environment. The utility and functionality is illustrated along a real world application scenario at the Ministry of Finance in Austria.
Cited by:	25
Resources:	[PDF] from researchgate.net

:view less

How do users express goals on the web?-an exploration of intentional structures in web search

2007

Markus Strohmaier, Mathias Lux, Michael Granitzer, Peter Scheir, Sotirios Liaskos, Eric Yu
International Conference on Web Information Systems Engineering, 67-78
:read more

Authors:	Markus Strohmaier, Mathias Lux, Michael Granitzer, Peter Scheir, Sotirios Liaskos, Eric Yu
Publication date:	12/03/2007
Journal:	International Conference on Web Information Systems Engineering
Pages:	67-78
Publisher:	Springer Berlin Heidelberg
Description:	Many activities on the web are driven by high-level goals of users, such as “plan a trip” or “buy some product”. In this paper, we are interested in exploring the role and structure of users’ goals in web search. We want to gain insights into how users express goals, and how their goals can be represented in a semi-formal way. This paper presents results from an exploratory study that focused on analyzing selected search sessions from a search engine log. In a detailed example, we demonstrate how goal-oriented search can be represented and understood as a traversal of goal graphs. Finally, we provide some ideas on how to construct large-scale goal graphs in a semi-algorithmic, collaborative way. We conclude with a description of a series of challenges that we consider to be important for future research.
Cited by:	31
Resources:	[PDF] from researchgate.net

:view less

Applying vector space models to ontology link type suggestion

2007

Albert Weichselbraun, Michael Granitzer, Gerhard Wohlgenannt, Thomas Neidhart, Arno Sch…
2007 Innovations in Information Technologies (IIT), 566-570
:read more

Authors:	Albert Weichselbraun, Michael Granitzer, Gerhard Wohlgenannt, Thomas Neidhart, Arno Scharl, Andreas Juffinger
Publication date:	11/18/2007
Journal:	2007 Innovations in Information Technologies (IIT)
Pages:	566-570
Publisher:	IEEE
Description:	The identification and labeling of non-hierarchical relations are among the most challenging tasks in ontology learning. This paper describes an approach for suggesting ontology relationship types to domain experts based on implicitly learned relations from a domain corpus. The learning process extracts verb-vectors from sentences containing domain concepts. It computes centroids for known relationship types and stores them in the knowledge base. Vectors of unknown relationships are compared to the stored centroids using the cosine similarity metric. The system then suggests the relationship type of the most similar centroid. Domain experts evaluate these suggestions to refine the knowledge base and constantly improve the component's accuracy. Using four sample ontologies on "energy sources", this paper demonstrates how link type suggestion aids the ontology design process. It also provides a …
Cited by:	3

:view less

Distributed Web2. 0 crawling for ontology evolution

2007

Andreas Juffinger, Thomas Neidhart, Albert Weichselbraun, Gerhard Wohlgenannt, Michael …
2007 2nd International Conference on Digital Information Management 2, 615-620
:read more

Authors:	Andreas Juffinger, Thomas Neidhart, Albert Weichselbraun, Gerhard Wohlgenannt, Michael Granitzer, Roman Kern, Arno Scharl
Publication date:	10/28/2007
Journal:	2007 2nd International Conference on Digital Information Management
Volume:	2
Pages:	615-620
Publisher:	IEEE
Description:	Semantic Web technologies in general and ontologybased approaches in particular are considered the foundation for the next generation of information services. While ontologies enable software agents to exchange knowledge and information in a standardised, intelligent manner, describing todays vast amount of information in terms of ontological knowledge and to track the evolution of such ontologies remains a challenge. In this paper we describe Web2.0 crawling for ontology evolution. The World Wide Web, or Web for short, is due, its evolutionary properties and social network characteristics a perfect fitting data source to evolve an ontology. The decentralised structure of the Internet, the huge amount of data and upcoming Web2.0 technologies arise several challenges for a crawling system. In this paper we present a distributed crawling system with standard browser integration. The proposed system is a …
Cited by:	14
Resources:	[PDF] from academia.edu

:view less

Evaluating an Information Retrieval System for the Semantic Desktop using Standard Measures from Information Retrieval.

2007

Peter Scheir, Michael Granitzer, Stefanie N Lindstaedt
LWA, 269-272
:read more

Authors:	Peter Scheir, Michael Granitzer, Stefanie N Lindstaedt
Publication date:	09/24/2007
Journal:	LWA
Pages:	269-272
Description:	Evaluation of information retrieval systems is a critical aspect of information retrieval research. New retrieval paradigms, as retrieval in the Semantic Web, present an additional challenge for system evaluation as no off-the-shelf test corpora for evaluation exist. This paper describes the approach taken to evaluate an information retrieval system built for the Semantic Desktop and demonstrates how standard measures from information retrieval research are employed for evaluation.
Cited by:	6
Resources:	[PDF] from researchgate.net

:view less

Aspects of broad folksonomies

2007

Mathias Lux, Michael Granitzer, Roman Kern
18th International Workshop on Database and Expert Systems Applications (DEXA 2007), 28…
:read more

Authors:	Mathias Lux, Michael Granitzer, Roman Kern
Publication date:	09/03/2007
Journal:	18th International Workshop on Database and Expert Systems Applications (DEXA 2007)
Pages:	283-287
Publisher:	IEEE
Description:	Folksonomies, collaboratively created sets of metadata, are becoming more and more important for organising information and knowledge of communites in the Web. While for a single user the difference to keyword assignment is marginal, the power of folksonomies emerges from the collaborative aspects. Folksonomies are already issue of research. Within this publication we analyse underlying statistical properties of broad folksonomies aiming to identify laws and characteristics, which allow inferring properties for folksonomy based retrieval. The actual benefit of folksonomies for retrieval and the derived methods are concluded from experiments with aggregated data from del.icio.us1.
Cited by:	53
Resources:	[PDF] from academia.edu

:view less

Fused Exploration of Temporal Developments and Topical Relationships in Heterogeneous Data Sets

2007

Vedran Sabol, Michael Granitzer, Wolfgang Kienreich
2007 11th International Conference Information Visualization (IV'07), 369-375
:read more

Authors:	Vedran Sabol, Michael Granitzer, Wolfgang Kienreich
Publication date:	07/04/2007
Journal:	2007 11th International Conference Information Visualization (IV'07)
Pages:	369-375
Publisher:	IEEE
Description:	In many application areas which deal with heterogeneous data sets temporal developments and topical relationships both play an important role. A number of visual representations have been developed which separately address each of these two aspects of the data. However, a simultaneous analysis of both aspects is also often required. In this paper we introduce an approach which utilizes a combination of visual representations, where some are targeting the temporal behaviour while others are focusing on topical relationships. The views are fused in the sense that they are tightly coupled with respect to the visualised data set, considered time interval, and user actions. The resulting user interface enables the user to simultaneously explore and discover patterns in both temporal and topical aspects of a heterogeneous data set. As a proof of concept a demo application was developed and applied for visualising …
Cited by:	6

:view less

Automated ontology learning and validation using hypothesis testing

2007

Michael Granitzer, Arno Scharl, Albert Weichselbraun, Thomas Neidhart, Andreas Juffinge…
Advances in Intelligent Web Mastering: Proceedings of the 5th Atlantic Web Intelligence…
:read more

Authors:	Michael Granitzer, Arno Scharl, Albert Weichselbraun, Thomas Neidhart, Andreas Juffinger, Gerhard Wohlgenannt
Publication date:	06/15/2007
Journal:	Advances in Intelligent Web Mastering: Proceedings of the 5th Atlantic Web Intelligence Conference–AWIC’2007, Fontainbleau, France, June 25–27, 2007
Pages:	130-135
Publisher:	Springer Berlin Heidelberg
Description:	Semantic Web technologies in general and ontology-based approaches in particular are considered the foundation for the next generation of information services. While ontologies enable software agents to exchange knowledge and information in a standardized, intelligent manner, describing todaydocumentclass[12pt]{minimal} usepackage{amsmath} usepackage{wasysym} usepackage{amsfonts} usepackage{amssymb} usepackage{amsbsy} usepackage{mathrsfs} usepackage{upgreek} setlength{oddsidemargin}{-69pt} begin{document}$check{rm S}$end{document}s vast amount of information in terms of ontological knowledge remains a challenge. In this paper we describe the research project AVALON - Acquisition and VALidation of ONtologies, which aims at reducing the knowledge acquisition bottleneck by using methods from ontology learning in the context of a cybernetic …
Cited by:	7
Resources:	[HTML] from academia.edu

:view less

Visualization Metaphors for Multi-modal Meeting

2007

Vedran Sabol, Christian Gütl, Thomas Hermann Neidhart, Andreas Juffinger, Hans Werner K…
Workshop on Multimedia Semantics, 999-999
:read more

Authors:	Vedran Sabol, Christian Gütl, Thomas Hermann Neidhart, Andreas Juffinger, Hans Werner Klieber, Michael Granitzer
Publication date:	01/01/2007
Journal:	Workshop on Multimedia Semantics
Pages:	999-999
Publisher:	.
Description:	Visualization Metaphors for Multi-modal Meeting - Graz University of Technology Skip to main navigation Skip to search Skip to main content Graz University of Technology Home Graz University of Technology Logo English Deutsch Search content at Graz University of Technology Home Persons Publications Organisations Projects Activities Prizes Press/Media Visualization Metaphors for Multi-modal Meeting Vedran Sabol , Christian Gütl , Thomas Hermann Neidhart , Andreas Juffinger , Hans Werner Klieber , Michael Granitzer Institute of Human-Centred Computing (7060) Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review Overview Original language English Title of host publication Proc. Band "Aachener Informatik Berichte" Publisher . Pages 999-999 Publication status Published - 2007 Event Workshop on Multimedia Semantics - Aachen, Germany Duration: 5 Jun …

:view less

Special Track on Knowledge Organization and Semantic Technologies-Introduction

2007

Johannes Liegl, Mathias Lux, Viktoria Pammer, Peter Scheir, Michael Granitzer, Stefanie…
International Conference on Knowledge Management, 330-331
:read more

Authors:	Johannes Liegl, Mathias Lux, Viktoria Pammer, Peter Scheir, Michael Granitzer, Stefanie Lindstaedt
Publication date:	01/01/2007
Journal:	International Conference on Knowledge Management
Pages:	330-331
Publisher:	.
Description:	Special Track on Knowledge Organization and Semantic Technologies - Introduction - Technische Universität Graz Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln Technische Universität Graz Home Technische Universität Graz Logo English Deutsch Home Personen Publikationen Organisationen Projekte Aktivitäten Preise Presseberichte Nach Expertise, Namen oder Zugehörigkeit suchen Special Track on Knowledge Organization and Semantic Technologies - Introduction Johannes Liegl, Mathias Lux, Viktoria Pammer, Peter Scheir, Michael Granitzer, Stefanie Lindstaedt Publikation: Beitrag in Buch/Bericht/Konferenzband › Vorwort/Nachwort Übersicht Originalsprache englisch Titel International Conference on Knowledge Management Herausgeber (Verlag) . Seiten 330-331 Publikationsstatus Veröffentlicht - 2007 Dieses zitieren APA Standard Harvard Vancouver Author BIBTEX …

:view less

Task Instance Classification via Graph Kernels

2007

Mark Kröll, Andreas Rath, Nicolas Weber, Stefanie Lindstaedt, Michael Granitzer
Mining and Learning with Graphs, Italy
:read more

Authors:	Mark Kröll, Andreas Rath, Nicolas Weber, Stefanie Lindstaedt, Michael Granitzer
Publication date:	01/01/2007
Journal:	Mining and Learning with Graphs, Italy
Description:	Knowledge-intensive work plays an increasingly important role in organisations of all types. Knowledge workers contribute their effort to achieve a common purpose; they are part of (business) processes. Workflow Management Systems support them during their daily work, featuring guidance and providing intelligent resource delivery. However, the emergence of richly structured, heterogeneous datasets requires a reassessment of existing mining techniques which do not take possible relations between individual instances into account. Neglecting these relations might lead to inappropriate conclusions about the data. In order to uphold the support quality of knowledge workers, the application of mining methods, that consider structure information rather than content information, is necessary. In the scope of the research project DYONIPOS, user interaction patterns, eg, relations between users, resources and tasks, are mapped in the form of graphs. We utilize graph kernels to exploit structural information and apply Support Vector Machines to classify task instances to task models.
Cited by:	5
Resources:	[PDF] from researchgate.net

:view less

International Workshop on Collaborative Knowledge Management for Web Information Systems (WEKnow)-How Do Users Express Goals on the Web?–An Exploration of Intentional …

2007

M Strohmaier, M Lux, M Granitzer, P Scheir, S Liaskos, E Yu
Lecture Notes in Computer Science 4832, 67-78
:read more

Authors:	M Strohmaier, M Lux, M Granitzer, P Scheir, S Liaskos, E Yu
Publication date:	01/01/2007
Journal:	Lecture Notes in Computer Science
Volume:	4832
Pages:	67-78
Publisher:	Berlin: Springer-Verlag, 1973-

:view less

:back to overview

Michael Granitzer

Publications