https://drive.google.com/file/d/1uYLqLsrMRgPd6SR1TzyI_SnW3PEC4SMH/view?usp=drive_link

Large Language Models (LLMs) provide new opportunities for AI-assisted ontology engineering, however they often perform poorly, creating complex class hierarchies, inconsistent taxonomies, uncontrolled property growth, and modelling anti-patterns.

This workshop introduces a practical hands-on methodology for AI-assisted ontology engineering, showing how LLMs can be transformed from unreliable ontology generators into effective, controlled modelling assistants within professional ontology development workflows.

Participants will learn a practical methodology for AI-assisted ontology engineering that introduces guardrails, modelling patterns, and structured workflows to guide LLM behaviour and prevent hallucinations and poor ontology design. The approach emphasises using LLMs to augment human ontology designers and engineering practices rather than generate ontologies automatically.

Hands-on exercises will guide attendees in integrating LLMs into their ontology development lifecycle:

configuring an ontology modelling toolchain
extending LLM capabilities with ontology-specific prompts, skills and agents
applying modelling guardrails and style guides
generating and evolving ontology structures with LLM support
auditing ontology quality
producing conformance metrics and reports
generating test harnesses and data
preparing ontologies for production use
Prerequisites: familiarity with ontologies, an LLM (Claude recommended), an ontology editor, a laptop with network access.