The 24th International Conference on Knowledge Engineering and Knowledge Management (EKAW-24) took place in Amsterdam, the Netherlands. Topics of EKAW24 included semantic web, knowledge management, knowledge discovery, information integration, natural language processing, intelligent systems, e-business, ehealth, humanities, and cultural heritage. In addition, this year's conference invited research articles focusing on algorithms, tools, methodologies, and applications that leverage the interplay between knowledge and Language Models. The Joint Proceedings of Posters, Demos, Workshops, and Tutorials of EKAW 2024 group together contributions to the Posters and Demos and Workshops and Tutorials tracks of the conference.
¶ nEvelop-O ‡eXtraction and eXploitation of long-TAIL Knowledge with LLMs and KGs (X-TAIL) CEUR
ceur-ws.org
The 24th International Conference on Knowledge Engineering and Knowledge Management (EKAW 2024) took place at the Amsterdam Science Park campus, Netherlands, from November 26th to November 28th, 2024. This volume contains the proceedings of the Poster and Demo Track, along with the Workshops and Tutorials that were co-located with the main conference.
EKAW (pronounced /ˈiːkoʊ/) serves as the premier forum where researchers and practitioners advance the fields of knowledge engineering and knowledge management. The conference attracts a diverse community of scholars, knowledge engineers, IT architects, and industry professionals who are shaping the future of knowledge-based systems and applications.
This year’s special theme, ”Knowledge in the Age of Language Models,” reflects the field’s evolution and its engagement with emerging technologies. The conference specifically sought contributions exploring the interplay between traditional knowledge engineering approaches and modern language models, aiming to deepen our understanding of how these technologies can complement and enhance each other.
The Posters & Demos Track complements the main conference by providing a forum for researchers to present late-breaking results, ongoing projects, and prototypes in knowledge engineering and management. This track ofers an interactive setting that facilitates discussion between presenters and attendees, allowing researchers to receive constructive feedback and explore potential collaborations.
This year, in alignment with EKAW 2024’s special theme “Knowledge in the Age of Language Models”, we received submissions across a broad spectrum of topics, from foundational knowledge engineering approaches to cutting-edge applications of language models in knowledge management. After a thorough peer review process, 27 high-quality submissions (11 demos and 16 posters) were accepted, spanning several key research areas. Among these contributions, “Alter Heritage: a Web App to Gather Expert Knowledge on Inclusive Cultural Heritage Metadata” was awarded Best Poster/Demo for its innovative approach to enhancing cultural heritage documentation through knowledge engineering principles. Notable mentions were also awarded to “LINTEXT”, “PODIO”, “QuerIA”, and the “Collaborative RDF Benchmark Suite”, each exemplifying excellence in diferent aspects of knowledge engineering and management, from visual knowledge modeling to language model applications.
The accepted submissions spanned several key research areas:
• Language model-enhanced knowledge engineering tools and methodologies • Ontology evaluation and entity typing using LLMs • Knowledge-augmented approaches for text generation and question answering
• Pattern-based ontology transformation and validation • Dynamic ontology serialization and mapping • Collaborative knowledge engineering tools • Knowledge extraction from historical and archival documents
• Cultural heritage metadata management • Patent classification and analysis • Nuclear industry knowledge management • Railway topology information systems • Digital humanities and archival processing
• Knowledge graph querying and visualization • Event relation extraction • Named entity recognition • Table-to-knowledge graph matching
The accepted submissions demonstrate both technical innovation and practical applicability, with many works featuring functional prototypes and demonstrations. Each submission was evaluated by at least two program committee members based on originality, significance, and relevance to EKAW’s research topics. Particular attention was paid to ensuring that all contributions, especially those involving language models, established clear connections to knowledge engineering and management principles.
We are especially pleased to note the diversity of approaches presented, from traditional knowledge engineering methodologies to novel applications leveraging large language models, reflecting the evolving landscape of our field while maintaining its core focus on efective knowledge representation and management.
Inna Novalija is a senior researcher at the Artificial Intelligence Laboratory at Jožef Stefan Institute, Slovenia. Her research interests focus on knowledge management, text mining and artificial intelligence applications in various domains. She has extensive experience in European research projects and has contributed to the development of innovative knowledge-based solutions for digital transformation. Carlos Badenes-Olmedo is a senior researcher at the Ontology Engineering Group (OEG) at Universidad Politécnica de Madrid, Spain. His research centers on knowledge graphs, natural language processing and machine learning, with particular emphasis on their applications in digital humanities and cultural heritage. He has contributed to numerous projects combining semantic technologies with artificial intelligence approaches.
Together, they brought their complementary expertise in knowledge engineering and artificial intelligence to curate an engaging and diverse Posters and Demos track that showcases the latest innovations in the field.
We sincerely thank the members of the Program Committee for their valuable contributions in reviewing the submissions and ensuring the high quality of the accepted papers: • Abdul Sittar (Josef Stefan Institute, Slovenia) • Andrea Cimmino Arriaga (Universidad Politécnica de Madrid, Spain) • Elvira Amador-Domínguez (Universidad Politécnica de Madrid, Spain) • Erik Novak (Jožef Stefan Institute, Slovenia) • Janez Brank (Jozef Stefan Institute, Slovenia) • Jože Rožanec (Jožef Stefan Institute, Slovenia) • Laura Hollink (Centrum Wiskunde & Informatica, Netherlands) • María Navas-Loro (Universidad Politécnica de Madrid, Spain) • Marieke van Erp (KNAW Humanities Cluster, Netherlands) • Mathias Zinnen (Friedrich Alexander Universität Erlangen-Nürnberg, Germany) • Marco Antonio Stranisci (University of Turin, Italy) • Nandana Mihindukulasooriya (IBM Research AI, USA) • Oleksandra Topal (Institut ”Jožef Stefan”, Slovenia) • Pablo Calleja (Universidad Politécnica de Madrid, Spain) • Patricia Martín-Chozas (Universidad Politécnica de Madrid, Spain) • Soto Montalvo (Universidad Rey Juan Carlos, Spain) • Valerio Basile (University of Turin, Italy) • Vincent Christlein (University Erlangen-Nuremberg, Germany)
Special recognition goes to our track chairs Laura Hollink (Centrum Wiskunde & Informatica) and Marieke van Erp (KNAW Humanities Cluster) for their dedication and leadership in coordinating the review process.
The following contributions were accepted for presentation in the Posters and Demos Track:
• PODIO: A Political Discourse Ontology (Ibai Guillen-Pacho, Ana Iglesias-Molina, Carlos Badenes
Olmedo and Oscar Corcho) • OWL Data Properties Ontologically (Vojtěch Svátek, Kateřina Haniková and Ondřej Zamazal) • Leveraging Meta-Modelling Language for Ontology Structuring and Validation (Zekeri Adams,
Martin Homola, Ján Kl’Uka and Vojtech Svatek) • An ontology classifying residues from the bioeconomy (Kim Schmidt, Kai Sven Radtke and Marco
Selig)
• QuerIA: Contextual Learning-Driven Questionnaire Generation and Assessment based on Large
Language Models (Paul Eyzaguirre Barreda and Carlos Badenes Olmedo) • Automatic Ontology Term Typing by LLMs: the impact of prompt and ontology variation (Upal
Bhattacharya, Maaike de Boer and Sergey Sosnovsky) • Enhanced LLM with ontologies for smart Knowledge Management in nuclear industry (Frédéric
Godest, Mouna El Alaoui, Victor Richet and Malhomme Olivier ) • Evaluating Large Language Model Literature Reviews in Interdisciplinary Science: A Systems
Biology Perspective (Charvi Jain, Sahar Vahdati, Nandu Gopan, Ivo F. Sbalzarini and Jens Lehmann)
• Alter Heritage: a Web App to Gather Expert Knowledge on Inclusive Cultural Heritage Metadata (Andrei Nesterov, Laura Hollink and Jacco van Ossenbruggen) • LINTEXT: A Visual Tool for Exploring and Modeling Knowledge in Text Documents (Riley
Capshaw and Eva Blomqvist) • TRENDy: a tool for temporal modelling and database generation (Stephan Maree, Richard Taylor and C. Maria Keet) • PatOMat2: A Tool for Pattern-Based Ontology Transformation using SPARQL (Ondřej Zamazal,
Martin Ledvinka and Vojtěch Svátek) • Visual Data and Schema Queries over Knowledge Graphs (Sergejs Rikacovs and Kārlis Čerāns) • EU Contract Hub: Towards a more accessible public procurement (Virginia Ramón-Ferrer, Álvaro
Fontecha, Carlos Badenes-Olmedo and Oscar Corcho) • Taxonomy for Patent Classification: A Step Towards Intelligent Patent Analysis ( Elham Motamedi,
Inna Novalija and Luis Rei) • Intelligent Fault Diagnosis of Cyber Physical Systems using Knowledge Graphs (Ameneh
Naghdipour, Benno Kruit, Jieying Chen, Peter Kruizinga, Godfried Webers and Stefan Schlobach) • LLM based chatbot to find Railway Topology and Rail Vehicle Information in Europe ( Mohammed
Rasheed and Marina Aguado)
• Named Entity Recognition for digitised archival documents in German (Nele Garay, Mahsa Vafaie and Harald Sack) • Streamlining Event Relation Extraction: A Pipeline Leveraging Pretrained and Large Language
Models for Inference (Gustavo Flores Miguel, Youssra Rebboud, Pasquale Lisena and Raphael Troncy) • Unlocking historical knowledge: a semantic web approach to medieval notarial document analysis (Ángel García-Menéndez, José Emilio Labra-Gayo and Daniel Gayo-Avello) • On the Role of Preprocessing on Matching Tables to Knowledge Graphs (Vishvapalsinhji Parmar,
Achraf Hadder and Alsayed Algergawy) Knowledge Management and Infrastructure • Realizing a Collaborative RDF Benchmark Suite in Practice (Piotr Sowiński and Maria Ganzha) • RDF2JSON-OM: Dynamic ontology serialization using ontology mapping paths (Patrik Kompuš ) • Welcome, newborn entity! On handling newly generated entities in ontology transformation (Vojtěch Svátek, Ondřej Zamazal, Kateřina Haniková, David Chudán, Mohammad Javad Saeedizade and Eva Blomqvist) • Towards Synthesizing E-Mail Conversations as Part of Knowledge Work Datasets with Large
Language Models (Desiree Heim, Christian Jilek, Adrian Ulges and Andreas Dengel)
The 24th International Conference on Knowledge Engineering and Knowledge Management (EKAW-24) encompasses the diverse realms of eliciting, acquiring, modeling, and managing knowledge. The conference addresses the pivotal role of knowledge in constructing systems and services for the semantic web, knowledge management, knowledge discovery, information integration, natural language processing, intelligent systems, e-business, e-health, humanities, cultural heritage, and beyond. In particular, EKAW-24 emphasises the role of Language Models (LLMs) in Knowledge Engineering and Management. Besides the regular conference tracks, EKAW also hosts workshops and tutorials. • EKAW Workshops provide an informal setting in which participants have the opportunity to discuss specific topics in an atmosphere that fosters the exchange of ideas via networking activities. • EKAW Tutorials enable attendees to fully engage in emerging methods and technologies relevant to knowledge representation and management, established schools of thought, and exciting application areas.
Topics for this year reflected the general scope of EKAW 2024, encouraging the following themes: • Emergent technologies and methods in knowledge acquisition or management. • New application domains or ways to rethink established ones. • Novel problems in knowledge acquisition. • Domain-oriented, interdisciplinary, or “blue-sky” research that may challenge established KR paradigms.
• The topic has a dedicated and expanding research community.
The ESWC-24 edition included eight parallel events: four Tutorials and four Workshops. The following tutorials were hosted: • Conversational Knowledge Capture Using the KNOW Ontology • MUHAI Tutorial: Enabling Meaning and Understanding in Human-centric AI • Tutorial on Creating and Accessing Knowledge Graphs for Action Parameterisation • Semantic Knowledge Modeling - Ontologies & Vocabularies These proceedings include the papers presented at the four EKAW24 workshops: • First Workshop on Knowledge Management for Numerical Modeling, Measurement & Simulation (KNUMS) – see Section 4 • eXtraction and eXploitation of long-TAIL Knowledge with LLMs and KGs (X-TAIL) 5 • The First Workshop on Evaluation of Language Models in Knowledge Engineering (ELMKE) – see Section 6 • First Workshop on Knowledge Management in Newsrooms (K-MIN) – see Section 7
Enrico Daga is a Senior Research Fellow at the Knowledge Media Institute (KMi) of The Open University in the UK. He investigates the application of knowledge graph technologies in data-intensive, sociotechnical environments: cultural heritage, smart cities and robotics, and healthcare. His current focus is developing novel knowledge graph construction and curation methods (SPARQL Anything) and promoting knowledge graphs to support frontier humanities and cultural engagement applications. Lise Stork is an assistant professor at the Intelligent Data Engineering Lab (INDELab) of the University of Amsterdam (UvA). She investigates knowledge engineering strategies for scholarly applications, with a focus on human-centric AI, applying her work in scholarly domains such as sociology, natural history and healthcare. Her current focus is on leveraging multi-modal scholarly data for knowledge discovery, and human-AI workflows for knowledge engineering, with a focus on eliciting tacit scientific knowledge.
Together, they brought their complementary expertise in knowledge engineering and artificial intelligence to organize an extremely successful day of satellite events, with more than a hundred participants engaged in learning and debating the latest innovations in the field. 4. First Workshop on Knowledge Management for Numerical
Modeling, Measurement & Simulation (KNUMS) Research on data and knowledge management has made much progress in recent years, culminating in best practices, widely-adopted standards and (commercial) systems for a wide range of applications. However, despite these advancements, many fields still face challenges in managing the knowledge about complex measurements and numerical models. These challenges span across a broad spectrum of disciplines, from engineering and natural sciences to sociology and economics, and extend to a variety of applications, including manufacturing, agriculture, medicine and public policy. In a wide range of domains, it remains dificult to keep track of the connection between measurement datasets and the physical processes that generate them, or the simulations that use them.
This workshop brought together a diverse set of perspectives from diferent traditions and attempted to establish common ground for how these various kinds of representation and processes might be integrated.
The first Workshop on Knowledge Management for Numerical Modeling, Measurement & Simulation (KNUMS) was opened by invited keynote Hans Onvlee (ASML Research), and featured 5 paper presentations. • Benno Kruit, VU Amsterdam, Netherlands • João Moreira, University of Twente, Netherlands • Victoria Degeler, University of Amsterdam, Netherlands • Henderik A. Proper, TU Wien, Austria • Cornelis Bouter, TNO, Netherlands • Margherita Martorana, VU University, Amsterdam • Flavio Pileggi, University of Technology Sydney, Australia • Jan-Christoph Kalo, University of Amsterdam, Netherlands • Dilek Dustegor, University of Groningen, Netherlands 5. eXtraction and eXploitation of long-TAIL Knowledge with LLMs and KGs (X-TAIL) Large Language Models (LLMs) store extensive knowledge within their parameters, easily accessible through natural language interaction. However, they struggle when probed for long-tail knowledge (information rarely encountered during training). Conversely, Knowledge Graphs (KGs) excel at structuring specialized information but are often incomplete. Leveraging the parametric knowledge of LLMs and the authoritative knowledge stored in KGs could advance long-tail knowledge extraction and enhance its exploitation.
The first edition of “X-TAIL, eXtraction and eXploitation of long-TAIL knowledge” aims to attract researchers and practitioners operating at the intersection of KGs and Generative AI. X-TAIL ofers an opportunity to engage in interdisciplinary discussions focusing on non-standard sources or working on methods and tools designed to aid in such scenarios.
The workshops was opened by a keynote of Jan-Cristophe Kalo (UBA) with title: What do Large Language Models ”know” about the World?. • Arianna Graciotti (University of Bologna) • Alba Morales Tirado (The Open University) • Valentina Presutti (University of Bologna) • Enrico Motta (The Open University) Programme Committee • Aldo Gangemi, University of Bologna, Italy • Andrea Schimmenti, University of Bologna, Italy • Andrea Zugarini, expert.ai, Italy • Angelo Salatino, Open University, UK • Antonello Meloni, University of Cagliari, Italy • Benno Kruit, Vrije Universiteit Amsterdam • Bohui Zhang, King’s College London, UK • Célian Ringwald, Inria Université Côte d’Azur, I3S, CNRS • Chiara di Bonaventura, King’s College London, UK • Delfina Sol Martinez Pandiani, Centrum Wiskunde & Informatica, Netherlands • Diego Reforgiato, University of Cagliari, Italy • Gianmarco Pappacoda, University of Bologna, Italy • Harald Sack, FIZ Karlsruhe, Germany • Jan-Christoph Kalo, University of Amsterdam, Netherlands • Mahsa Vafaie, FIZ Karlsruhe, Germany • Nicolas Lazzari, University of Pisa/University of Bologna, Italy • Rocco Tripodi, University of Venice, Italy • Stefano De Giorgis, CNR Catania/University of Bologna, Italy • Tabea Tietz, FIZ Karlsruhe, Germany
• Named Entity Recognition in Historical Italian: The Case of Giacomo Leopardi’s Zibaldone.
(Cristian Santini, Laura Melosi, and Emanuele Frontoni) • Evaluation of LLMs on Long-tail Entity Linking in Historical Documents. (Marta Boscariol, Luana
Bulla, Lia Draetta, Beatrice Fiumanò, Emanuele Lenzi, and Leonardo Piano) • Constrained Information Retrieval for Long-Tail Knowledge Extraction. (Nicolas Lazzari, Arianna
Graciotti, and Valentina Presutti) 6. The First Workshop on Evaluation of Language Models in
Knowledge Engineering (ELMKE) Language models (LMs) have been considered promising in numerous knowledge engineering (KE) tasks, such as knowledge extraction, knowledge base construction, and curation. However, their adoption introduces new challenges for evaluation. The assessment of LM-generated results remains limited, lacking a comprehensive and formally defined framework, and relies heavily on human efort, making it dificult to compare methods and reproduce experiments.
The ELMKE workshop and its future series aim to address this critical gap by spearheading a community-driven efort to automate and standardize evaluation. It seeks to unite expertise, perspectives, and pioneering works to advance novel paradigms for evaluating LMs in KE. This workshop will showcase innovative and published papers focusing on evaluation methods for diverse KE tasks, such as completion and generative tasks. Additionally, discussions will explore challenges related to transparency, human evaluation, and broader reflections on the implications of evaluation methods. By establishing plans for platforms and dashboards for collaborative work, participants and the community can contribute to the design and implementation of robust evaluation methods and benchmarks, fostering targeted discussions and long-term collaboration.
The workshop opened with a keynote by Dr. Valentina Tamma from the University of Liverpool, titled “Ontology Engineering Revisited: The Rise of the LLMs”. • Bohui Zhang (King’s College London) • Yuan He (University of Oxford) • Reham Alharbi (University of Liverpool) 7. First Workshop on Knowledge Management in Newsrooms (K-MIN) Newsrooms handle information with various sources and formats. Processing this heterogeneous information and making sense out of it is key to news production and dissemination steps. With the proliferation of Large Language Models, there is renewed interest and scope in applying knowledge representations to efectively handle and utilize such diverse information. The synergy between LLMs and knowledge graphs, for example, ofer potential to build knowledge representations from textual data in a retrieval-friendly way, while minimising hallucinations or factually-incorrect responses. Integrating such techniques in the newsroom workflow, however, remains challenging. Discussions about AI in journalism often focus on the automated generation of news reports to various capacities, whereas its application to knowledge, the most fundamental tool in a newsroom, is often overlooked.
In this exciting backdrop of emerging solutions, the first edition of the workshop on “Knowledge Management in Newsrooms” presented a platform for researchers working with the knowledge representations and management in the domain of journalism and featured 4 papers (listed below).
• Dr. Reshmi Gopalakrishna Pillai (Vrije University Amsterdam) • Dr. Laurence Dierickx (University of Bergen / Université Libre de Bruxelles) • Dr. Hannes Cools, University of Amsterdam • Prof. Dr. Antske Fokkens, Vrije University Amsterdam • Dr. Marc Gallofré Ocaña, University of Bergen • Dr. Leo Leppänen, University of Helsinki • Prof. Dr. Carl-Gustav Lindén, University of Bergen • Dr. Silvia Majo Vazquez, Vrije University of Amsterdam • Nicolas Mattis, Vrije University of Amsterdam • Dr. Subhayan Mukerjee, National University Singapore • Dr. Valeria Resendez, Vrije University of Amsterdam • Dr. Theresa Seipp, University of Amsterdam • Dr. Nadja Shaetz, University of Hamburg • Stefanie Sirén-Heikel, University of Helsinki • Prof. Dr. Wouter van Atteveldt, Vrije University Amsterdam • Prof. Dr. Arjen van Dalen, University of Southern Denmark