Ontology refinement made easier using embeddings and large language models
Ontology Refinement with Embeddings and LLMs: A Case Study on CASTON
Artificial Intelligence
Summary
Building reliable knowledge maps called ontologies is hard because mistakes can hide deep in the class hierarchies. The authors worked with a casting industry ontology named CASTON and showed a method that combines special mathematical representations (embeddings) and large language models to find and fix those mistakes. Their tested approach found a few key mistakes and proposed fixes that experts all agreed were good. This method is lightweight and can scale to larger ontologies.
What this means in practice
- •For ontology engineers: Improve the accuracy of large domain ontologies by automatically detecting and suggesting fixes for subtle class hierarchy errors.
- •For industrial knowledge base teams: Maintain and refine specialized industrial ontologies like CASTON to ensure consistency and reliability in digital knowledge models.
Tested on one dataset.
Authors
Hui Yang, Ziyu Li, Xuan Liu, Anders Jarfors, He Tan, Jiaoyan Chen
Abstract
Ontologies are widely used to represent and organize complex knowledge in domains such as biomedicine and healthcare. However, constructing reliable ontologies remains challenging, as class hierarchies, whether manually or automatically constructed, may contain subtle subsumption errors that require careful refinement. We study this problem in CASTON, a casting-domain ontology, and present a lightweight but scalable refinement workflow that combines a fine-tuned ontology embedding model with Large Language Models (LLMs). The embedding model is used to extract a limited number of candidate axioms for refinement together with possible replacement candidates, and an LLM then returns repair suggestions. In a small expert evaluation, the default configuration using a CASTON-fine-tuned OnT embedding model and Qwen3-8B produced reliable high-confidence repairs: all six high-confidence replacements were accepted by both domain experts. The implementation is available at https://github.com/HuiYang1997/OntRefine.