LLMs for Ontology Engineering

Core Contribution

Examination of how Large Language Models can assist with ontology creation, extension, and maintenance. Includes methods for LLM-assisted concept extraction, relationship identification, and quality checking.

Key Patterns

  • LLM-generated ontology candidates + human validation
  • Consistency checking using LLM reasoning
  • Gap identification through LLM analysis of domain text

Challenges

  • Hallucination in relationship extraction
  • Inconsistent terminology generation
  • Need for domain expert oversight

Related: 06-molecule—knowledge-graph-construction, 06-molecule—llm-hallucination-knowledge-engineering