Enhancing Ontology Engineering with LLMs: From Search to Active Learning Extensions

Ganna Kholmska, Klemen Kenda and Jože Rožanec

Abstract
This paper explores the application of LLMs to ontology
engineering tasks within the HumAIne project. The authors
present a methodology for for LLM-driven discovery, analysis,
and extension of ontologies related to Data Mining, Machine
Learning, and manufacturing domains. LLMs were employed to
identify suitable ontologies and extend them with Active
Learning concepts. While LLMs significantly streamline
ontology-related tasks, challenges remain, including the need for
domain-specific validation in pilot case studies.