CAOS | Emsusto | FOAM | FOMI | FOUST | IFOW | KM4LAW | OK4I | OntoCom | PwM | Semantic Shields | ST4DM | Workshop on the Convergence of Large Language Models and Ontologies | YODA
and for the workshops (CAOS) (Emsusto) K. Kutt, J. T. F. Breis, A. Pichler, G. Palko´, G. J. Nalepa (YODA) ´I. Oliveira, D. F. Santamaria, T. v. Ede, D. Klein, G. Engelberg, G. Castiglione, G. Bella, A. Continella, G. Guizzardi (Semantic Shields) A. G. Castro, W. Suess, T. Hubauer (OK4I)
(Early Career Symposium) (Ontology Showcase)
(Demonstrations) (Research and Industrial Project Exhibition) (Journal-First) (Tutorials) https://www.utwente.nl/en/eemcs/fois2024
These proceedings collect the papers and extended abstracts presented in the Joint Ontology Workshops - JOWO 2024: Episode X: The Tukker Zomer of Ontology.
JOWO is the main venue of the International Association for Ontology and its Applications (IAOA) for workshops on formal and applied ontology. Since 2023, the event has been held with the IAOA’s flagship conference Formal Ontology in Information Systems (FOIS).
JOWO is an umbrella event that promotes interdisciplinary research and international collaboration. As such, it welcomes workshops covering a wide range of disciplines, including, but not limited to, Cognitive Science, Knowledge Representation, Conceptual Modeling, Robotics, Natural Language Processing, Artificial Intelligence, Logic, Philosophy, and Linguistics. With this interdisciplinary approach, JOWO seeks to provide a platform for the diverse communities interested in developing, reasoning with, and applying formalised ontologies. Since 2015, each edition of JOWO has its own character, with a diferent set of workshops and tutorials selected by the annual organizing team that reflects the respective local research communities and global research trends. As an overarching event covering all aspects of the IAOA community, the Joint Ontology Workshops (JOWO) continue to grow in importance and influence, reaching its tenth-anniversary milestone this year.
JOWO X was hosted by the University of Twente, the Netherlands, between 1517 July 2024. The event ofered a vibrant program featuring 14 workshops that spanned a broad spectrum of formal ontology research, addressing both foundational aspects (FOUST VIII) and applications in diverse areas, such as in cognitive science (CAOS VIII), sustainability (EMSUSTO), knowledge and data managment (FOAM and ST4DM), industries (FOMI and OK4I), modelling the food industries (IFOW 2024), law (KM4LAW), conceptual modeling (ONTOCOM X), Cybersecurity (Semantic Shields I), Large Language Models (Workshop on the Convergence of Large Language Models and Ontologies), and digital archives (YODA). Additionally, the program ofered also less traditional research endeavours, such as exploring interaction design techniques for learning, developing, and using ontologies and ontology-based conceptual models with Playing with Meanings (PwM workshop).
The complete list of workshops that were included in the tenth edition of JOWO is as follows: CAOS: Cognition And OntologieS Workshop EMSUSTO: Energy, Materials and Sustainability Ontology Workshop FOAM: FAIR Principles for Ontologies and Medatata in Knowledge Management
Workshop FOMI: Formal Ontologies Meet Industry Workshop FOUST: Workshop on Foundational Ontology IFOW: The Integrated Food Ontology Workshop KM4LAW: 3rd International Workshop - Knowledge Management and Process
Mining for Law OK4I: Ontologies and Knowledge Graphs for Industry ONTOCOM: International Workshop on Ontologies and Conceptual Modeling PmW: Playing with Meanings Workshop Semantic Shields: 1st International Workshop on Modeling for Cybersecurity ST4DM: Semantic Technologies for Data Management Workshop
Language Models and Ontologies YODA: ContemporarY Ontologies for Digital Archives Workshop Across the 14 workshops, 78 contributions were presented at JOWO X. The scientific program was further enriched by several keynote talks and panel discussions within the workshops.
We would like to thank all authors and speakers for their contributions, and the programme committee members and additional reviewers for their timely reviewing. We would also like to extend our gratitude to the local organizers at the University of Twente for making JOWO X such a remarkable event, and the International Association for Ontology and its Applications (IAOA) for providing organisational support.
Claudenir M. Fonseca Guendalina Righetti
´Italo Oliveira Pedro Paulo F. Barcelos Rodrigo Calhau University of Twente, The Netherlands University of Oslo, Norway
Stefano Borgo Oliver Kutz Frank Loebe Fabian Neuhaus Laboratory for Applied Ontology (ISTC-CNR), Italy Free University of Bozen-Bolzano, Italy University of Leipzig, Germany Otto von Guericke University Magdeburg, Germany
Stefano De Giorgis Guendalina Righetti Gabriele Sacco Maria M. Hedblom Oliver Kutz Greta Adamo Cristina Amoretti Daniel Beßler Roberta Ferrario Bart Gajderowicz Laura Giordano Torsten Hahmann Martha Lewis Todd Oakley Alessandro Oltramari Mihai Pomarlan Daniele Porello Kai Sauerwald Marco Schorlemmer He Tan
National Research Council (ISTC-CNR), Italy University of Oslo, Norway Free University of Bozen-Bolzano, Italy Jo¨nko¨ping University, Sweden Free University of Bozen-Bolzano, Italy
Basque Center for Climate Change, Spain University of Genoa, Italy University Bremen, Germany Naitonal Research Council (CNR), Italy University of Toronto, Canada Universiat` del Piemonte Orientale, Italy University of Maine, US University of Bristol, UK Case Western Reserve University, US Bosch Research and Technology Center, US University of Bremen, Germany University of Genoa, Italy University of Hagen, Germany Artificial Intelligence Research Institute (IIIA), Spain
Jo¨nko¨ping University, Sweden The core goal of ‘CAOS: Cognition And OntologieS’ is to investigate the relationship between cognition and ontologies to model, simulate, and represent cognitive phenomena for artificial intelligence.
Its interdisciplinary nature makes it the perfect environment to bridge the gap between knowledge representation, cognitive sciences, and formal ontology approaches. It provides a platform for researchers in either domain to discuss and present their work.
With this in mind, the primary focus of the workshop series lies in the formal modelling and representation of important cognitive phenomena and concepts, encompassing notions from research on language, reasoning, and behaviour. The event attracts an interdisciplinary audience from diverse fields, such as philosophy, linguistics, psychology, cognitive science, computer science, and other related disciplines.
We are delighted to announce that the eighth edition of CAOS received papers covering a wide range of topics, contributed by experienced researchers and students from diferent domains. In this edition, we accepted six papers for publication in this volume. The accepted contributions reflect the interdisciplinary nature of the workshop. Some of the works are focused on the relationship between cognitive approaches and ontological modelling for specific domains, as it is in the case of “From Human Cognitive Expertise to Ontological Formalization: Bridging the Knowledge Gap for Nanophotonic Calculator Design and Simulation” by Ouassila Labbani Narsis, Erik Dujardin, Christophe Nicolle, and Nicolas Gros, which investigates the conjunction of cognitive psychology and physics. “What is Abstraction in Biomimetics?” by Ludger Jansen focuses instead on the wedding between cognitively inspired ontological modelling and biology, in particular biological mimicry and its relation with foundational ontologies. At the crossroads with the medical domain, we can locate “Understanding ASD: Design and Development of a Domain Ontology to Assist Professionals In Understanding Autistic Children Based on DSM5”, where Jessica Lima, Victor Oliveira, Patıcr´io Silva, and Sebasti˜ao Alves Filho propose the development of a domain ontology for the domain of autism, with the scope of assisting professional during the diagnosis.
Physics, language, and ontological modelling are the domains explored in “The Geography of Temperature Space” by George Wright and Matthew Purver, who pave the way to represent ontologically the geometrical space for temperaturerelated terminology.
In a similar line of investigation about cognitive approaches to ontological modelling and language, Maria Keet’s “Preliminary steps toward an ontology for noun classes in Niger-Congo languages” ofers a relevant debate about the potentiality of shaping reality depending on the language and the inner bias in adopting one or another. Finally, the paper“Ontologies, Arguments, and Large-Language Models”, by John Beverley, Francesco Franda, Hedi Karray, Daniel Maxwell, CarterBeau Benson, and Barry Smith, explores the intersection of ontologies, argumentation, and Large Language Models by extending ARGO, the Arguments Ontology, to promote justification and traceability of LLM.
To conclude this edition of CAOS, we had the pleasure of hosting Prof. Laura Palmigiano as the keynote speaker of the event. The keynote captured the audience by discussing ”Categories and Categorization”, and by providing examples of how methods, insights, and techniques about structural proof theory, algebraic logic, duality theory, and category theory in mathematics can be used in synergy with one another to develop an overarching logical theory of categories and categorization which can be used to model typicality and related cognitive phenomena.
John Beverley Mark Jensen Dan Maxwell Hedi Karray Paul Kogut
University at Bufalo, USA Customs and Border Protection, USA
KaDSci, LLC, USA University of Technology of Tarbes, France
Lockheed Martin, USA The rapid deployment of Large-Language Models (LLMs) has led to recognition of both their value and limitations. It is unnecessary at this point to populate the literature with more examples of LLM ’hallucinations’. More important is the parallel recognition by researchers exploring this novel terrain that certain LLM limitations can be addressed by leveraging knowledge graphs and ontologies. In retrospect, this should have been no surprise. One should not expect architectures based largely on statistical methods to reliably produce trustworthy outputs or indeed provide transparent explanations for when outputs difer from expectations. Expectations aside, it has long been recognized that ontologies and knowledge graphs grounded in well-developed formal languages such as the Web Ontology Language (OWL), can provide precisely such transparent explanations, i.e. formal proofs, and if set up correctly, provide reliably trustworthy output as well. We are at present amid an emerging research area - across academia, governments, and industry - at the intersection of LLMs and knowledge representation, as groups from various quarters aim to push the boundaries of artificial intelligence. The central purpose of this workshop was to further that push, as evidenced by organizing/program committee members, invited talks, and peer-reviewed presentations, boasting representatives from academia, government, and industry. The collaborative spirit was, moreover, magnified by the association of this workshop with a special issue of the journal Applied Ontology, dedicated to exploring the convergence of knowledge representation and LLM strategies, design patterns, models, and benchmarks. Submissions were received from both the JOWO EasyChair portal and the Applied Ontology journal portal to be considered for CLLMO. We were pleased to receive for this first edition of CLLMO 16 such submissions, of which 8 were selected for the workshop, with 6 published in this volume.
The workshop itself was divided into four major interest areas: framing convergence, convergence on explainability, convergence on application, and convergence on ethics. The program, slides, and recorded talks can be found on the workshop web page. Topics ranged from exploring ontological models underwriting LLM outputs and supporting explainability for LLM output using knowledge graphs, to promoting ontology learning via vector embeddings and supplementing LLMs with ontology-based Retrieval Augmented Generation (RAG). Our workshop was fortunate enough to have three distinguished keynotes. Barry Smith, from the University at Bufalo, delivered a provocative talk on the creative limitations of artificial intelligence, which led naturally into our second keynote, by Jeremy Ravenel of NAAS, who emphasized methods for building trust in artificial intelligence systems. Our last keynote, Beth Rudden of Bast AI, demonstrated a full stack workflow, leveraging LLMs and knowledge representation throughout, aimed at promoting transparency in articfiial intelligence. Given the substantial interest this workshop generated, we added a post-session section to the program web page. There you will find recorded talks, papers, and slides covering convergence topics which - for various reasons - were not able to be included in the workshop proper.
Martin Glauer Mirjam Stappel
Otto von Guericke University Magdeburg, Germany Fraunhofer Institute for Energy Economics and Energy System Technology (IEE), Germany
Fabian Neuhaus Jo¨rg Waitelonis Carsten Hoyer-Klick
Otto von Guericke University Magdeburg, Germany FIZ Karlsruhe - Leibniz Institute for Information Infrastructure, Germany
German Aerospace Center (DLR), Germany The Energy, Materials, and Sustainability Ontology (Emsusto) Workshop 2024 provided a venue for presenting ontologies and their applications in the domains of renewable energy, materials, and sustainability. The primary goal of this workshop was to ofer researchers and practitioners in these domains an opportunity to meet and exchange experiences in developing and using ontologies and to provide a forum for the exchange of knowledge and best practices for ontologies in these domains.
The second installation of this workshop focussed on the heterogeneous landscape of energy ontologies. Various aspects of these diferent ontologies were introduced in the presentations. One of the key challenges emphasised in the presentations was that there is a multitude of parallel ontology developments revolving around energy, its conversion, transport, and use. This is particularly problematic as the ontological representation of energy is notoriously dificult. Diferent approaches to modelling such a central concept present an insurmountable obstacle to the integration and cross-use of the resulting ontologies. The following discussion focussed on how the future development of these previously independent resources can be better combined. In the course of this debate, other experts were also involved, allowing additional energy-related ontologies to be included in this convergence process.
Cassia Trojahn Maıar´ Poveda-Villaol´n Daniel Garijo Ces´ar Henrique Bernabe´ Asiyah Yu Lin
Institut de Recherche en Informatique de Toulouse, France Ontology Engineering Group, Universidad Polietc´nica de Madrid, Spain Ontology Engineering Group, Universidad Polietc´nica de Madrid, Spain Leiden University Medical Center, the Netherlands
Axle Research and Technology, USA
Nathalie Aussenac-Gilles IRIT CNRS, France Luiz Bonino University of Twente, The Netherlands Maya Daneva University of Twente, The Netherlands Alban Gaignard CNRS, France Giancarlo Guizzardi University of Twente, The Netherlands Clement Jonquet MISTEA (INRAE) and LIRMM, France Wanderley Lopes de Souza Universidade Federal de S˜ao Carlos (UFSCar), Brazil Barbara Magagna GO FAIR Foundation, The Netherlands Anna-Maria Masci University of Texas MD Anderson Cancer Center, USA Jo˜ao Moreira Univesitait Twente, The Netherlands Peter Mutschke Leibniz Institute for the Social Sciences, Germany Robert Pergl Czech Technical University, Czech Republic Katy Wolstencroft Leiden University, The Netherlands Veruska Zamborlini Federal University of Espirito Santo, Brazil The FAIR principles have been proposed by the scientific community to provide guidance on making research outputs more Findable, Accessible, Interoperable, and Reusable on the Web. The FAIR principles have gained increasing attention in a number of diferent domains and applications. On the one hand, a key aspect promoted by FAIR is the ability to describe resources correctly and semantically, in particular using ontologies or vocabularies. On the other hand, ontologies themselves have to comply with the FAIR principles.
The main objectives of the ‘FAIR Principles for Ontologies and Medatata in Knowledge Management’ (FOAM) workshop are 1) to bring together new methods for enabling FAIR research outputs (data, software, workflows, ontologies, alignments, etc.) through ontologies and vocabularies; 2) to discuss techniques, metrics and guidelines to improve the FAIRness of ontologies and vocabularies; and 3) to share experiences, identify new challenges and opportunities applying FAIR for diferent communities.
FOAM is a joint initiative resulting from merging the Workshop on Ontologies for FAIR and FAIR Ontologies (Onto4FAIR), the Workshop for Conceptual Modeling, Ontologies and Metadata Management for Findable, Accessible, Interoperable and Reusable (FAIR) Data (CMOMM4FAIR) and the FAIR Ontology Harmonization and TRUST Data Interoperability Workshop (FOHTI). The first edition of FOAM the workshop accepted five full publications and a poster (as a separate invited presentation). All submissions were subject to peer review, with a minimum of 2 reviews per paper. The presented works can be grouped into three main categories. The rfist category dealt with FAIR aspects of semantic artefacts: Poveda-Villaol´n et. al. described a gap analysis in ontology engineering methodologies when making vocabularies FAIR by design, while Marınt´ kova´ et al described the challenges associated with creating FAIR crosswalks and mappings. Magagna et al. provided an analysis of diferent resources supporting the FAIR principles.
The second category of publications explored application-specific requirements. El Ghosh et al addressed semantic interoperability in the oncology domain, while Hannou et al. explored the use of agile methodologies in adapting FAIR for European Data Spaces.
Finally, the invited presentation described the FAIRness assessment of semantic artefacts in the environmental domain.
The workshop ended with a discussion session where the following key points were highlighted: 1. The existence of diferent tools for evaluating diferent aspects of FAIRness of semantic artefacts and the need for an alignment between them. 2. The need for common practices across domains on how to make data/artefacts
FAIR. 3. The need for tutorials and lectures on how to construct (good) ontologies and how to reuse ontologies across diferent domains.
lifecycle as highlighted in Poveda-Villaol´n et. al. work.
invited talk, include a tutorial/hands-on, and a project networking session (as several international projects in the topic have addressed these topics). These points open several opportunities and motivate future editions of FOAM. 13th International Workshop on Formal Ontologies meet Industry
Francesco Compagno Anna Fensel Walter Terkaj Universiat` di Trento, Italy Wageningen University and Research, Netherlands UNational Research Council of Italy (CNR), Italy
Bahar Aameri University of Toronto, Canada Farhad Ameri Arizona State University, United States Rebeca Arista Airbus, France Silvia Chiacchiera STFC/UKRI, United Kingdom Ana Correia ATB, Bremen, Germany Laura Daniele TNO, Netherlandss Peter Klein Fraunhofer ITWM, Germany Serm Kulvatunyou SID-NIST, United States Riichiro Mizoguchi Japan Advanced Institute of Science and Technology, Japan Elisa Negri Politecnico di Milano, Italy Alessandro Oltramari Bosch Research and Technology Center, United States Herve´ Panetto University of Lorraine (CNR), France Pieter Pauwels Eindhoven University of Technology, Netherlands Maıar´ Poveda-Villaol´n Universidad Polietc´nica de Madrid, Spain Emilio M. Sanfilippo National Research Council of Italy (CNR), Italy Arkopaul Sarkar ENIT, France Lorenzo Solano Universitat Polietc`nica de Vaeln`cia, Spain Daniele Spoladore CNR-STIIMA, Italy Alessandro Umbrico National Research Council of Italy (CNR), Italy Marcela Vegetti INGAR (CONICET/UTN), Argentina Laure Vieu Institut de Recherche en Informatique de Toulouse, France Francesco Antonio Zaccarini University of Bologna, Italy FOMI is an international forum where academic researchers and industrial practitioners meet to analyse and discuss application issues related to methods, theories, tools, and applications based on formal ontologies. There is today a wide agreement that knowledge modelling and the semantic dimension of information play an increasingly central role in the networked economy: semantic-based applications aim to provide a framework for information and knowledge sharing, reliable information exchange, meaning negotiation and coordination between distinct organizations or among members of the same organization. The 13th edition of FOMI saw a significant attendance (more than 35 people), welcoming academic researchers and industry practitioners. This event provided a platform for open discussions on shared challenges and solutions related to ontology applications, without restrictions on the domains involved. The accepted contributions at FOMI 2024, eight in total, tackle heterogeneous topics. Four works are focused on specific industrial domains. Two works deal with the Oil & Gas domain, considering data retrieval tasks (Rodrigues et al.) and interoperability of specifications (Petry et al.). Additionally, Skrzek et al. analyze manufacturing in the aerospace domain, while Kitamura et al. investigate patents related to inorganic materials.
The remaining four works concentrate on ontology-based methodologies with a broader scope. Hurley et al. present the development of an ontology-based foundation for digital twins, while Moreira et al. introduce a framework to facilitate ontology reuse by software engineers. Finally, two papers are related to ontology alignment, one in a general context (Zaccarini et al.) and the other focused on the life-science domain (De Colle et al.).
Salvatore Florio
University of Oslo, Norway
Speaker : Nicola Guarino Riccardo Baratella Riichiro Mizoguchi Frank Loebe Moderator: Giancarlo Guizzardi
Laboratory for Applied Ontology, ISTC-CNR, Trento, Italy University of Genoa, Italy Japan Advanced Institute of Science and Technology, Japan University of Leipzig, Germany
University of Twente, The Netherlands Foundational ontologies are attempts to systematize very high-level categories of what exists. In practical terms, they aim to understand and formalise the meaning of general ontological entities, such as physical and abstract object, relations, property, event, and process. Once formalised, these entities provide the top-level categories that are, in principle, common to many (if not all) application domains. Consequently, foundational ontologies are acknowledged as tools for tackling systems interoperability and explainability issues. The Workshop on Foundational Ontology (FOUST) aims to create a platform for researchers to explore the foundational aspects of applied ontology. This encompasses discussing philosophical foundations, presenting new research on specific foundational ontologies, discussing existing foundational ontologies, comparing them, and examining their relevance to the broader ontological enterprise. The 8th workshop edition, FOUST VIII, added a special topic to attract attention to the role of constructional approaches in applied ontology. The concept of constructional ontology, introduced by Kit Fine in their 1991 influential paper The Study of Ontology, involves assuming a set of ontological givens or basic elements and a set of constructors that can be iteratively applied to generate new elements in the ontology. Elements of the ontology emerge through the constructional process. The idea is related to the iterative conception of sets (Go¨del 1964, Boolos 1971), which has been proven to be a powerful and elegant paradigm in the philosophy of mathematics to prevent paradoxes (e.g., the “Russell set”). FOUST VIII served as a platform to kickstart the discussion on how to leverage similar ideas in applied ontology. The workshop took place in two days, featuring a keynote, a panel, and ten paper presentations. On the first day, FOUST VIII focused on processes, events, and temporal entities. In the first session, Baratella presented a novel reconciliatory approach to the Puzzle of Uniqueness within perdurantism. The work of Rodrigues, Carbonera, and Abel proposed an analysis and modelling of events from a system-based perspective. The vibrant panel “From Events to Processes and Back,” between Nicola Guarino, Riccardo Baratella, Riichiro Mizoguchi, and Frank Loebe, moderated by Giancarlo Guizzardi, wrapped up the first day by entertaining the public with an engaging, interactive debate. The session captivated the audience by addressing the most salient challenges and emerging trends in the ontological modelling of processes and events.
The second day of the workshop revolved around constructional approaches together with more traditional foundational topics. It began with the keynote of Salvatore Florio on “Introduction to Constructional Ontology,” ofering a comprehensive perspective on constructional approaches, the discussion of the possible modelling choices such approaches can ofer, and their values for ontologies. The constructional contributions comprised the investigation of Berneri, bringing together classical subjects in formal ontology, i.e., metaphysical dependence, with constructional angles. Partridge, Mitchell, de Cesare, Cola, Khan, Price, and Hierl presented the BORO case history as a concrete example of exploiting the constructional approach in applied ontology. Finally, Righetti applied constructional ideas to the notion of concept. The last session of the workshop was dedicated to most traditional topics. The work of Gnatenko, Kutz, and Troquard (winner of the Best of JOWO award) proposed an ontologically motivated knowledge base for computational complexity theory designed to store, query, and reason over the extensive body of knowledge on algorithmic problems, complexity classes, and the like. The research of Porello, Vieu, Borgo, Compagno, Sanfilippo, and Terkaj presented a core OWL2 module of the well-known top-level ontology DOLCE. The study of Yargan and Jansen proposed an investigation of the definitions of functions reflecting on their appropriateness in the biomimetics context. Finally, two papers by Toyoshima, Barton, Koslicki, and Massin focused on the notion of artefacts and artifactual functions, ofering a realisable-oriented approach.
Damion Dooley Michaela Ku¨mpel Georgeta Bordea Robert Warren Anoosha Sehar Matthew Lange Felix Bindt Ido Toxopeus The Integrated Food Ontology Workshop at FOIS/JOWO 2024 (University of Twente, Enschede, Netherlands) was a full day of twelve presentations organized under four topic areas: Food Ontology modelling, Farm to Fork Food System Representation, LLM/ChatGPT and its role in ontology development and database/ knowledge graph search, and Food Safety. The event facilitated a discussion of many diferent leading-edge areas of research touching on food robotics, Chinese philosophy about food and health, ontological coding of food processing contamination risks, ChatGPT interface to food composition databases, and more!
Davide Audrito Francesca Grasso Roberto Nai Emilio Sulis Legal Studies Department, University of Bologna, Italy Computer Science Department, University of Turin, Italy Computer Science Department, University of Turin, Italy Computer Science Department, University of Turin, Italy
Valerio Basile University of Turin, Italy Valerio Bellandi University of Milan, Italy Guido Boella University of Turin, Italy Chiara Di Francescomarino University of Trento, Italy Beatriz Esteves Ghent University, Belgium Marcelo Fantinato University of S˜ao Paulo, Brazil Rohan Nanda University of Maastricht, Netherlands Maıar´ Navas Loro Universidad Polietc´nica de Madrid, Spain ´Italo Joes´ da Silva Oliveira Free University of Bozen-Bolzano, Italy Monica Palmirani University of Bologna, Italy Harshvardhan J. Pandit Trinity College, Dublin Sergio Picascia University of Milan, Italy Davide Riva University of Milan, Italy Livio Robaldo Swansea University, Wales Vıc´tor Rodıgr´ uez-Doncel Universidad Polietc´nica de Madrid, Spain Elena Romanenko Free University of Bozen-Bolzano, Italy Massimiliano Ronzani FBK, Trento, Italy Salvatore Sapienza University of Bologna, Italy Galileo Sartor University of Bologna, Italy Giovanni Siragusa University of Turin, Italy Andrea Tagarelli University of Calabria, Italy Andrea Vandin Sant’Anna School for Advanced Studies, Pisa, Italy Artificial Intelligence (AI), Knowledge Modeling (KM), Information Extraction (IE) and Process Mining (PM) methods are becoming increasingly relevant to numerous sub-domains of legal informatics. These areas include ontologies, argumentation, natural language processing, and legal event log analysis, all of which can be paired with a multilingual approach. The Knowledge Management and Process Mining for Law (KM4Law) workshop is a forum to discuss these and other related topics.
The swift advancement of AI in recent years has brought us closer to solving long-standing challenges in AI & Law. This progress makes it all the more important to identify the limits of automated systems, especially when faced with the remaining unsolved intentional and unintentional ambiguities and conflicts that demand legal interpretation. This workshop aims to shed light on these issues, exploring the yet unfaced opportunities and challenges that AI presents for knowledge representation in the legal domain.
The goals of our workshop range widely, covering the classification of legal sources, legal design, and legal ontologies. Also included are legal decisions similarity and clustering, prediction, and support during judicial decision-making, and legal interpretation support. Further topics encompass identifying the evolution of legal concepts and definitions over time, information extraction and classification, process mining for legal compliance, and detecting linguistic phenomena and patterns in legal sources. We also focus on multilingual alignments of concepts, both domestic and international, and the identification of legal references and network analysis. In particular, the third edition of the international workshop KM4LAW featured a keynote about challenges in Legal Knowledge Representation by ´Italo Oliveira. Five papers were presented at the workshop, covering diverse topics such as an Ontology-based approach in a conversational system by Frida Milella et al.; a Cross-Country Study Using Topic Modeling on Legal Documents from India and the UK by Krish Didwania et al.; an ontology for legal reasoning on data sharing and processing between law enforcement agencies by Jeremy Bouche-Pillon et al.; Exploring Usability in Regulatory Information Extraction Process by Antoine Sacer´ et al.; a Multilingual Logical English for Double Taxation Conventions by Galileo Sartor et al. The event’s success and the diversity of topics discussed highlight the workshop’s relevance to current research in the field.
Alexander Garcia Castro Thomas Hubauer Wolfgang Suess Philipp Schmurr
Siemens Energy, Germany Siemens, Germany Karlsruhe Institute of Technology, Germany Karlsruhe Institute of Technology, Germany Although semantic technologies have made significant advances in the last years, adoption within organizations remains a key challenge in the industry. The goal of this workshop is to foster exchange on key challenges for adoption, and solutions to overcome them. We are interested not only in concrete use cases based on knowledge graphs and ontologies and how they are embedded in the overall enterprise environment, but also in visionary proposals on how recent improvements in the area of semantic technologies and generative AI can help improve data integration and management, knowledge discovery, and AI use cases. OK4I welcomes contributions describing positive as well as negative results, coming from small or large companies as well as academic partners. Across all submissions, emphasis should be put on demonstrating the business value and impact (intended to be) created by using knowledge graphs and semantic technologies to address industry problems.
Questions addressed by OK4I: • How are KGs and ontologies being adopted? What are industry examples of implementations of these technologies, and what did you do to ensure success? • What are the main dificulties when using KGs and ontologies? What are examples of a failed adoption, and which lessons have you learnt on this journey? • Ontologies and KGs in manufacturing, supply chain management, PLM, logistics, procurement, etc. • Ontologies supporting CAD interoperability and feature extraction; towards smart CAD environments. • KGs and ontology generation from textual data • How are KGs and ontologies being used in combination with LLMs? • How is AI (e.g. LLMs) being used to do KGs and ontology enrichment? • Where are KGs and ontologies in the Data Mesh, data contracts? How are ontologies and KGs a part of modern data architectures? • Practical cases of successful and unsuccessful application of ontologies and KG technologies in application domains such as financial, biomedical, e-business, engineering, law enforcement, document management, government, and legislative systems. • How are KGs and ontologies helping the realization of FAIR data? Key learnings from the presentations and the workshop include: • Ontology modeling is (classically) a collaborative process between ontologists and subject-matter experts, which requires finding a common language to bridge the gap between technology and domain. To ensure success, ontologists need to learn the ”language of the domain”, and SMEs need to embrace some rigidity in thinking required for structured data modeling. • Tooling for ontology development could be improved significantly to better support this collaborative process. • To ensure the success of knowledge-graph-based solutions across enterprises and reduce hurdles to adoption by users and developers outside the ”semantics bubble”, it is essential to integrate KG-based systems into the standard enterprise information infrastructure using APIs or microservices. This should not be limited to information access, but also include means to create and update information.
10th International Workshop on Ontologies and Conceptual Modeling
University of Westminster, UK Ghent University, Belgium University of Twente, The Netherlands University of Westminster, UK and BORO Solutions, UK Universidad Polietc´nica de Valencia, Spain
Jo˜ao Paulo Almeida Federal University of Espır´ito Santo, Brazil Mike Bennett Hypercube, UK Salvatore Florio University of Oslo, Norway Pierre Grenon National Center for Ontological Research, USA Renata Guizzardi University of Twente, Netherlands Paul Johannesson Royal Institute of Technology, Sweden Jim Logan Dassault Sysetm`es, USA Andrew Mitchell BORO Solutions, UK Thomas Moser St. Po¨lten University of Applied Sciences, Austria ´Italo Joes´ da Silva Oliveira Free University of Bozen-Bolzano, Italy Jefrey Parsons Memorial University of Newfoundland, Canada Geert Poels Ghent University, Belgium Tiago Prince Sales University of Twente, Netherlands Pnina Sofer University of Haifa, Israel Marzieh Talebpour University of Westminster, UK Karsten Tolle Goethe University Frankfurt, Germany This year the International Workshop on Ontologies and Conceptual Modeling (OntoCom X) celebrated its 10th edition.
The importance of conceptual modeling has grown over the years, and it is now common to find examples of conceptual models being developed and used in a range of diverse disciplines not related to computing, including, for example, biology, business, construction, and engineering. Among the reasons for this disciplinary expansion is also the increasing digitalization of all aspects of modern life, as well as the increased complexity that such digitalization entails in terms of emerging needs and requirements. The natural consequence is a proliferation of conceptual models of multiple real-world domains, which sooner or later require data and systems to interoperate and/or integrate. In this emerging scenario, ontology-driven conceptual modeling becomes even more fundamental to modern life due to its intrinsic ability to represent reality in a theoretically and semantically consistent manner. Foundational (or upper ontologies) have the potential to resolve the dificult problems that derive from a lack of a consistent and sound ontological theory. The benefits that can derive from the adoption of a foundational ontology include improved mapping to the real world domain, increased level of communication and understanding among stakeholders, model reuse, semantic integration and interoperability and increased overall eficiency and efectiveness of information systems development and evolution. The application of foundational ontologies can also assist in overcoming the inscrutable nature of most mainstream artificial intelligence methods (i.e. neural networks and machine learning).
The OntoCom workshop series aim to bring together academics, researchers and practitioners to develop an agenda of future collaborations that combine research and industrial expertise. The research papers presented at the workshop contribute toward this goal.
Contributions were sought in the form of research papers. The submissions were reviewed by three program committee members and evaluated by the organizing committee. Four papers were accepted. These papers focused on the following research areas: the ontological grounding of simulation modeling, the adoption of ontology to assist in the interoperability of legal systems, development of a robust ontological account of what capabilities are, and the definition of sound criteria for characterizing mid-level ontologies.
The workshop concluded with an inspiring keynote by Henderik A. Proper, Professor in Enterprise and Process Engineering in the Business Informatics Group at the TU Wien (Austria). His talk, titled “Understanding the Variety of Domain Models: Views, Programs, Animations, and Other Models”, focused on the foundations of modeling and how diverse modeling artifacts can be used within the context of model-driven systems development.
Finally, we thank all the authors for their valuable contributions, our keynote speaker, and our entire program committee for their insightful reviews and constructive feedback.
Greta Adamo Max Willis
Basque Center for Climate Change, Spain
Universitat Polietc`nica de Vaeln`cia, Spain The Playing with Meanings (PwM) workshop was proposed to explore emergent interaction design methodologies for ontology learning and co-design. Preliminary investigations of participatory sense-making techniques using ontologies, first outlined in [?], led to the design of several tangible games, which serve as discursive artefacts that apply ontologies to extend contemporary practices of stakeholder engagement and knowledge elicitation.
The goals of this workshop were to (i) reflect on the challenges of learning and applying ontological notions, (i) introduce the game, play, and participatory sensemaking concepts, (iii) engage participants hands-on with ontology-based games and collaborative modelling, and (iv) to gather feedback on ontology-based tangible artefacts as participatory research tools.
The PwM workshop was structured around group interactions focused on formal ontologies and ontology-based conceptual models. Workshop participants relfected through action on the unique dispositions of games and play to create a space of social learning, structure communication, and elucidate complexity in an accessible and fun way.
The workshop activities were introduced through brief presentations outlining the background material for the two workshop sessions. The first session focused on the foundational game artefact Type Token [?] which introduces several Unified Foundational Ontology (UFO) constructs [?]; the second deployed Risk Response [?], an adversarial discourse game that originated from the ontological analysis and unpacking of climate change risk [?]. The two sessions were independent, and participants could attend one session without being bound to the content of the other. A pre- and post-experience questionnaire was ofered to participants, who numbered 19 in the first session and between 14 and 19 in the second session. Type Token. The first PwM session began with an outline of the project’s origins and an introduction to participatory sense-making. Participants were then given the first of two decks of cards, and instructed in the rules of gameplay. Seated around three tables, two groups of 4 to 6 players, and one group of 10 players, began collaborative ontological modelling using the Type Token cards (see Fig. 1, left). After several rounds of gameplay, a second deck of cards was introduced, which when added to the first, allowed for more extensive modelling. The session was concluded with an activity debrief and discussion. Risk Response. The second session started with a presentation of some background interaction design knowledge pertinent to the forthcoming activity, and an introduction to the social-ecological systems integrated conceptual model [?] which would be used. This was followed by an embodied sense-making exercise in which the participants role-played elements of the integrated conceptual model to become acquainted with the model’s entities and relations. Participants were then familiarised with the Risk Response game, its origins, and rules and formed teams to propose, respond to, and debate plausible (and preposterous) future adaptation and mitigation, referring to a modified futures cone diagram [ ?]. As part of gameplay, participants engaged in collaborative modelling (see Fig. 1, right) of risk and responses, according to the reference integrated conceptual model, then debated whether the proposed solutions could be viable in a future scenario. In a second round, participants modelled their game interactions using snippets of the Common Ontology of Value and Risk (COVER) [?], a UFO-based ontology specifically dealing with, for example, risk-experience, object at risk, and threat event. The Risk Response session was concluded with a debrief, in which the activities and outcomes were discussed, which elaborated, for example, trade-ofs between model quality and the quality of discourse among players, and the roles of the conceptual models, ontologies, and collaborative modelling exercises to stimulate group discourse and foster sense-making.
University of Twente, The Netherlands Santa- University of Catania, Italy
University of Twente, The Netherlands Accenture Labs, Israel Accenture Labs and University of Haifa, Israel University of Catania, Italy University of Catania, Italy University of Twente, The Netherlands University of Twente, The Netherlands
University of Twente, The Netherlands University of Oxford, UK University of Twente, The Netherlands University of Quebec in Outaouais, Canada Santa- University of Catania, Italy
´Italo Oliveira Daniele Francesco maria Thijs van Ede Dan Klein Gal Engelberg Gianpietro Castiglione Giampaolo Bella Andrea Continella Giancarlo Guizzardi Semantic Shields I: 1st International Workshop on Modeling for Cybersecurity was a forum to discuss theoretical aspects and practical applications of conceptual models and ontologies to cybersecurity. Cybersecurity, which concerns human and technological aspects, refers to a set of techniques to protect the integrity of networks, programs, and data from attack, damage, or unauthorized access. With the spread of systems and applications, attacks continue to grow in sophistication, with attackers using an ever-expanding variety of tactics such as social engineering, malware, and ransomware. New methods and technologies have been emerging to reduce the risk of cyber-attacks and protect against exploitations. Formal methods, particularly ontologies, constitute an efective approach to mitigate the incompleteness and ambiguity of security directives and semantically characterize security stakeholders, ranging from ofensive techniques to compliance, vulnerability, encryption, data protection, authentication, confidentiality, integrity, and availability. The Semantic Shields aims to bring together cybersecurity experts, conceptual modelers, and ontologists, from scholars to practitioners, to develop applications, methods, and tools impacting the cybersecurity domain.
Davide Lanti Alessandro Mosca Federico Maria Scafoglieri Guohui Xiao Free University of Bozen-Bolzano, Italy Free University of Bozen-Bolzano, Italy University of Rome La Sapienza, Italy University of Bergen, Norway
Ahmet Soylu OsloMet – Oslo Metropolitan University, Norway Antonis Bikakis University College London, UK Roman Kontchakov Birkbeck University of London, UK Antonella Poggi La Sapienza University of Rome, Italy Jennifer D’Souza TIB - Leibniz Information Centre for Science and Technology University Library, Germany David Chaves-Fraga USC - Universidade de Santiago de Compostela, Spain Vladislav Ryzhikov Birkbeck University of London, UK Tarcisio Mendes de Farias Universiet´ de Lausanne, Switzerland / SIB Swiss Institute of Bioinformatics, Switzerland Oscar Corcho UPM - Universidad Polietc´nica de Madrid, Spain Gianluca Cima University of Rome La Sapienza, Italy Elena Botoeva University of Kent, UK Francesco Corcoglioniti Free University of Bozen-Bolzano, Italy Ognjen Savkovic Free University of Bozen-Bolzano, Italy Valentina Bartalesi Lenzi ISTI-CNR Pisa, Italy Tiago Prince Sales University of Twente, The Netherlands Jose M Parente de Oliveira Aeronautics Institute of Technology, Brazil ´Italo Da Silva Oliveira University of Twente, The Netherlands Jo˜ao Rebelo Moreira University of Twente, The Netherlands Anastasia Dimou KU Leuven, Belgium Filipi Miranda Soares University of Twente, The Netherlands Anisa Rula University of Brescia, Italy Luca Andolfi University of Rome La Sapienza, Italy Knowledge Graphs (KGs) have become a popular format of data representation mainly due to their flexible data model, which renders them particularly suited to those tasks where data coming from multiple, possibly heterogeneous, sources has to be integrated in order to be fully exploited. KGs have received both the attention of academia, through foundational eforts stemming from scientific research areas such as Knowledge Representation, Machine Learning, or Databases, and the enterprise world. Enterprise applications, in particular, exploit tools implementing standard recommendations from the Semantic Web community (such as RDF and OWL), and proprietary formats based on property graphs. The general data model of KGs allows for representing both extensional knowledge, the data itself, and intensional information made available by means of domain ontologies. Hence, KGs provide a way to enrich data coming from legacy sources with semantic information coming from the application domain and expert knowledge. This empowers users with automated inference support, enriches interpretability of data, and overall facilitates data access and integration.
The goal of this workshop is to create a specialized forum for researchers and practitioners working at the intersection of semantic technologies and KGs, focusing on both data and ontologies. Many existing venues tend to emphasize one or the other — such as the Database community’s focus on data, or the Knowledge Representation and Semantic Web communities’ focus on ontologies. However, initiatives focusing on leveraging semantic technologies for data management necessitate an integrated perspective that merges data with semantics. This also calls for bespoke techniques that seamlessly combine the two.
The first edition the ST4DM workshop opened with a keynote by Antonella Poggi, who discussed metamodeling (also known as “multi-level modeling”) and metaquerying for Data Management. This was followed by four presentations that spanned a range of subjects, from foundational research to practical applications in various fields. Specifically, the presentations covered: implementing controlled query evaluation in Ontology-Based Data Access systems by Divya Baura et al.; improving the cost of updates in Virtual Knowledge Graphs by Romuald Esdras Wandji et al.; Ontology-Driven Data Management Design in Health-Care Domain by Leonardo Cocco et al.; and integrating multiple Knowledge Graphs in Digital Humanities by Alberto Morvillo et al.
We trust that this workshop serves as a fruitful platform for advancing the integration of data and semantics through Knowledge Graphs, inspiring future work and collaboration in this dynamic area.
Krzysztof Kutt Jesualdo Tomas´ Fernan´dez Breis Alois Pichler Ga´bor Palko´ Grzegorz J. Nalepa Martin Atzmu¨ller Jesualdo Tomas´ Fernan´dez Breis Rune Falch Zos´fia Fellegi Rafael Valencia Garıac´ Jakub Gomu lka Christophe Guillotel-Nothmann Krzysztof Kutt Ulrich Lobis Grzegorz J. Nalepa Ga´bor Palko´ Joes´ Toma´s Palma Men´dez Alois Pichler Elz˙bieta Sroka Joseph Wang-Kathrein
Jagiellonian University, Poland University of Murcia, Spain
University of Bergen, Norway Eot¨vo¨s Loar´nd University of Budapest, Hungary
Jagiellonian University, Poland
Osnabur¨ck University, Germany University of Murcia, Spain
University of Bergen, Norway Institute for Literary Studies (HUN-REN), Hungary University of Murcia, Spain
AGH University of Krakow, Poland Centre National de la Recherche Scientifique, France Jagiellonian University, Poland University of Innsbruck, Austria
Jagiellonian University, Poland Eot¨vo¨s Loar´nd University, Hungary University of Murcia, Spain
University of Bergen, Norway Lukasiewicz – EMAG, Poland
University of Innsbruck, Austria The primary goal of the YODA workshop is to ofer an interdisciplinary venue for discussion of all issues related to the development of ontology-based systems for storing cultural heritage artifacts. Therefore, the workshop is open to submissions from all stakeholders working in any institution in the GLAM sector (galleries, libraries, archives, and museums) or using their resources, including collection curators, librarians, philologists, historians, enthusiasts, and knowledge engineers. A wide range of theoretical and practical papers are accepted, covering topics from crafting a model underlying a collection, through compiling metadata about individual documents and digital collections and considering integration across systems of various institutions to exchange and complement information about collections, to supporting advanced research processing scenarios. The YODA workshop was packed into a single session with five papers presented, each of which was accepted based on the evaluation of at least 2 reviewers. It started with two conceptual presentations.
In the first, the authors considered how to properly model texts and works. Since there is no consensus here even among scholars, the authors proposed a new approach that relies on design patterns instead of a monolithic ontology, which will be more robust. The second conceptual presentation was a kind of tutorial on how to choose the vocabularies that best suit the goals and needs of a specific project, from among a whole range of diferent vocabularies to describe archival documents. The second part of our adventure focused on presentations of three use cases. It started with the Wittgenstein Ontology, which will model the philosophical content of Ludwig Wittgenstein’s writings digitized in the Wittgenstein Archives at the University of Bergen, Norway. Then, we saw the Our Heritage, Our Story project, which with the help of natural language processing techniques seeks to connect community archives from across the United Kingdom. In the last presentation, we looked at attempts to create an ontology to describe manuscripts stored at the Jagiellonian University, Poland.
Zubeida C. Khan Council for Scientific and Industrial Research, South Africa Pedro Paulo Favato Barcelos University of Twente, Netherlands, and Ghent University, Belgium Jo˜ao Moreira Univesitait Twente, The Netherlands The Early Career Symposium (ECS) is an integral part of the FOIS conference, created with the specific aim of providing students and early career researchers with a stage to present their work and receive feedback and insights from experienced researchers.
In the spirit of cultivating young talent and fostering knowledge exchange, this year’s participants in the ECS had the chance to showcase their research through both short oral presentations in front of the FOIS audience and engaging poster presentations.
Each participant could engage in an extended conversation with a senior mentor in their respective efilds during a dedicated mentoring lunch. This interaction serves as a bridge between generations of researchers, facilitating guidance, wisdom, and knowledge transfer from experienced hands to those just starting their academic journey, and, in turn, contributes to the collective advancement of knowledge by allowing young researchers to provide innovative ideas.
The ECS is not just about formal presentations and structured mentoring but also ofers a venue to build informal networks and relationships. To facilitate this, we hosted an ECS breakfast, creating a relaxed and friendly atmosphere where early career researchers and their mentors could interact with each other more informally, exchange ideas, and build lasting connections.
As it is connected to the FOIS conference, the ECS welcomes research addressed in an interdisciplinary aptitude towards formal and philosophical ontology, cognitive science, knowledge representation, linguistics, and more.
We express our deepest gratitude to all participants, mentors, and attendees for making ECS an essential component of the FOIS conference.
Laura Daniele Jo˜ao Paulo A. Almeida
TNO, The Netherlands Federal University of Espır´ito Santo, Brazil As the Applied Ontology community, we have reached the point where an impressive variety of ontologies have been developed across a wide range of domains. For the most part, however, there has been a lack of coordination among these eforts and even a lack of awareness about the work that is being done by groups within the community. The Ontology Showcase at FOIS 2024 facilitates the sharing and reuse of ontologies, to achieve the vision of seamless semantic interoperability of curated ontologies within their applications.
Pawel Garbacz Robert Pergl
John Paul II Catholic University of Lublin, Poland Czech Technical University in Prague, Czech Republic
University of Twente, Netherlands King Abdullah University of Science and Technology, Saudi Arabia Bosch Research and Technology Center, USA Czech Technical University in Prague, Czech Republic
CNR-STIIMA, Italy The demonstration track complements FOIS 2024 main tracks by ofering an interactive platform for authors to present and discuss their work. It invites demonstrations of methods and tools developed using ontologies and those to create, maintain, integrate, publish, evaluate, and implement ontologies. It also welcomes demonstrations of novel ontology (anti)patterns and challenges arising in the ontology engineering life cycle.
This year, five papers were accepted in the demonstration track. One was presented during the on-site part of FOIS and the remainder in the online part. The papers discussed tools for education, reasoning systems, and the application of knowledge graphs.
Enrico Franconi Paul Johannesson Free University of Bozen-Bolzano, Italy Stockholms Universitet, Sweden
Adrien Barton Emilio M. Sanfilippo
Institut de Recherche en Informatique de Toulouse (IRIT), CNRS, France
Laboratory for Applied Ontology ISTC-CNR, Italy For the rfist time within the FOIS context, the Journal-First Track provided a forum for discussing recently published research on topics related to formal ontology, aiming to facilitate broader dissemination within the FOIS community. This track invited submissions of extended abstracts from papers that had been published or accepted for publication in journals, provided they had not been previously presented at FOIS or any other major venue focused on applied ontology, such as ICBO, ISWC, ESWC, K-CAP, or JOWO workshops.
Free University of Bozen-Bolzano, Italy
Federal University of Espır´ito Santo, Brazil The tutorial sessions of the FOIS 2024 edition aimed to provide attendees with an engaging learning experience, balancing both theoretical exploration and practical applications in key areas of ontology and knowledge engineering. Through expert-led discussions and interactive exercises, four tutorials ofered participants opportunities to delve into discussions on ontological analysis of events and processes, the engineering of hybrid intelligent systems, and advanced methods for ontology implementation in OWL. 1. “Knowledge Engineering for Hybrid Intelligence (CommonKADS in Hybrid Scenarios)” by Ilaria Tiddi, Victor de Boer, and Stefan Schlobach from Vrije Universiteit Amsterdam, introduced the CommonKADS methodology in hybrid intelligence contexts, combining explanation of the methology with handson activities. 2. “Events, Processes, and their Descriptions” by Nicola Guarino from ISTCCNR Trento, shared his expertise on the conceptualization of events and processes, shedding light on their descriptions in ontological terms. 3. “Generating Ontology Conceptualization and Pattern Libraries with Chowlk” by Maıar´ Poveda-Villaol´n, Rau´l Garıac´ -Castro, and Sergio Carulli-Per´ez from Universidad Polietc´nica de Madrid, was a hands-on tutorial introducing participants to Chowlk, a tool for generating ontology conceptualizations and pattern libraries through practical exercises. 4. “DOLCE in OWL” by Emilio M. Sanfilippo and Walter Terkaj from CNR ISTC and CNR STIIMA from Italy, provided a combined theoretical and practical session on implementing the DOLCE foundational ontology in OWL. We express our gratitude to the presenters for their invaluable contributions and to the participants who gained valuable insights into current challenges and methodologies, leaving them with enhanced skills and fresh perspectives to apply in their own research and projects.