The 23rd International Semantic Web Conference (ISWC) once again highlighted its role as the premier global venue for advancing semantic web and knowledge graph technologies. These innovations, essential for fostering interoperability and streamlining data enrichment, are closely aligned with the mission of the enRichMyData project. Naturally, our teams played a prominent role in this year’s event, contributing cutting-edge research and practical demonstrations to the field.
Research Spotlight: LLMs for Ontology Learning
The Bosch team and the University of Mannheim explored the role of Large Language Models (LLMs) in Ontology Learning, addressing whether LLMs can reason over unstructured data or rely on linguistic patterns. Findings revealed that fine-tuning significantly improves LLM performance, even with domain-specific terms, positioning them as valuable tools for Knowledge Base Completion (KBC).
Hands-On STI Tutorial
The UNIMIB team, alongside Ernesto Jiménez-Ruiz (University of London), delivered a tutorial on Semantic Table Interpretation (STI)—a critical process for mapping tabular data to knowledge graphs. Attendees gained insights into LLM-based approaches and engaged in hands-on training using entity linking tools.
Award-Winning Demo
The SemT-UI prototype, a tool for interactive tabular data enrichment, was recognized as a runner-up for the Best Demo Award. As part of the SemT framework, it enables flexible STI workflows and emphasizes open-source collaboration.
Congratulations to our teams for their remarkable contributions to advancing data enrichment and semantic technologies!
