KG4SDSE Workshop Description

The goal of this cross-over workshop is to stimulate convergence between the communities focusing on Knowledge Graphs (KGs) and those working on Model-driven Engineering of Information Systems. The initiative is motivated, among other factors, by the emerging notion of 'Context Graphs' highlighting the importance of traceable decision-making and process knowledge, something that has been traditionally captured through enterprise modeling languages (e.g. BPMN, DMN, CMMN, Archimate or domain-specific adaptations). Context design has been an on-going concern in enterprise modeling and systems engineering communities, typically enabled by diagrammatic methods that have gradually incorporated RDF graphs or OWL ontologies towards diverse application cases - KG generation, semantic integration or reasoning over diagrammatic content, traceability or enrichment of decisions and process designs, semantic contextualization for business process management systems etc. The recent uptake of the Agentic AI paradigm, with renewed interest in orchestration, delegation and semantic contextualization of agentic capabilities, further suggests a growing relevance of enterprise context representations and lineage.

Legacy enterprise model repositories are semantically underutilized, often treated as graphical documentation – e.g. of business processes or enterprise architecting decisions. Moving beyond this legacy perception, the Model-driven Engineering community developed expertise and methods that leverage graphical notations for machine interpretability, code generation, process automation, model evaluations. KGs add new possibilities, potentially transforming knowledge engineering processes, their input artifacts and outcomes. Furthermore, integration patterns between KGs and LLMs can benefit from model-driven engineering practices, towards novel knowledge-driven architectures.

Therefore, this workshop aims to stimulate a two-way exchange of knowledge and technical enablers - how artifacts and tools commonly employed in Model-driven Engineering can support KG -based systems or processes; and vice versa, we also look at how modeling methods and model content processing can be empowered by KGs and KG-based AI systems. Research advances on the interplay between KG-based neurosymbolic AI approaches and conceptual modeling methods are also of particular interest to our workshop, towards promoting new types of knowledge flows and knowledge-based system architectures.

Various stages of research maturity can be reported in the workshop contributions - from position and vision papers to experience reports and full research papers.

Relevant Topics

• Model-driven integration of KGs and LLMs
• Enterprise models completion, alignment and enrichment based on KG techniques
• Enterprise models for orchestration and delegation in Agentic AI
• Automated reasoning in Model-driven engineering
• KGs as mediators for metamodeling and LLMs
• Information Systems engineering methods based on KGs
• Visual syntaxes and low-code approaches to building KGs
• Model-driven Context Graphs and context engineering
• KG-based Enterprise Architecture and Business Process Management
• Requirements engineering, system design and analysis augmented by KGs
• KGs for model-based Digital Twins
• Linking and transforming domain-specific diagrammatic models with KGs
• Enriching KGs with domain-specific and enterprise models
• Empirical studies and experience reports on the interplay between model-driven engineering and KGs or neuro-symbolic AI

Submission Guidelines

Submission link coming soon!

We invite two types of paper contributions:

• Full papers which can be regular research or experience papers (9-12 pages including references)
• Short papers which can be position or vision papers (6-8 pages including references)

Since we plan to submit the accepted paper for inclusion in a CEUR-WS volume, formatting must comply with the CEURART 1-column format. Information about the CEURART format can be found at https://ceur-ws.org/HOWTOSUBMIT.html (scroll to section “CEURART style files for papers”). The initial PDF submission can use the deprecated Word template (http://ceur-ws.org/Vol-XXX/CEURART.zip) but camera-ready drafts of accepted papers will have to use current CEURART Latex/Overleaf templates (Overleaf template available at https://www.overleaf.com/latex/templates/template-for-submissions-to-ceur-workshop-proceedings-ceur-ws-dot-org/wqyfdgftmcfw; and offline Latex template available at http://ceur-ws.org/Vol-XXX/CEURART.zip).

Use of Generative AI in the development of the paper must comply with the CEUR-WS GenAI policy and must be explicitly declared at the end of the paper, as instructed at https://ceur-ws.org/GenAI/Policy.html.

All accepted papers will have to be presented in person by one registered author. Accepted papers that will not presented will be included the workshop proceedings.

Workshop Chairs

• Robert Buchmann (main contact: robert.buchmann@ubbcluj.ro), Babeș-Bolyai University, Romania
• Dimitris Karagiannis (dk@dke.univie.ac.at), University of Vienna, Austria
• Dimitris Plexousakis (dp@ics.forth.gr), University of Crete & FORTH, Greece

Web Presence Chair

Iulia Vaidian (iulia.vaidian@omilab.org), OMiLAB NPO, Germany

Workshop Program Committee (to be extended)

• Amin Anjomshoaa, Vienna University of Economics and Business, Austria
• Hans-Georg Fill, University of Fribourg, Switzerland
• Anne Füßl, Technische Universität Ilmenau, Germany
• Aurona Gerber, University of West Cape, South Africa
• Ana-Maria Ghiran, Babeș-Bolyai University, Romania
• Marite Kirikova, Riga Technical University, Latvia
• Jose Emilio Labra Gayo, Universidad de Oviedo, Chile
• Ana León, Universitat Politècnica de València, Spain
• Bruno Oliveira, Porto Polytechnic, Portugal
• Andreas Opdahl, University of Bergen, Norway
• Jeff Z. Pan, University of Edinburgh, UK
• Andrea Polini, University of Camerino, Italy
• Anisa Rula, University of Brescia, Italy
• Yannis Tzitzikas, University of Crete and ICS-FORTH, Greece

Contact

robert.buchmann@ubbcluj.ro