Complex Ontology Matching with Large Language Model Embeddings
This work addresses a specific challenge in knowledge graph matching for domains requiring expressive correspondences, representing an incremental improvement over existing embedding-based methods.
The paper tackled the problem of generating expressive correspondences in ontology and knowledge graph matching by integrating large language models (LLMs) into an approach based on alignment need and ABox-based relation discovery, resulting in a 45% increase in F-measure compared to baseline methods.
Ontology, and more broadly, Knowledge Graph Matching is a challenging task in which expressiveness has not been fully addressed. Despite the increasing use of embeddings and language models for this task, approaches for generating expressive correspondences still do not take full advantage of these models, in particular, large language models (LLMs). This paper proposes to integrate LLMs into an approach for generating expressive correspondences based on alignment need and ABox-based relation discovery. The generation of correspondences is performed by matching similar surroundings of instance sub-graphs. The integration of LLMs results in different architectural modifications, including label similarity, sub-graph matching, and entity matching. The performance word embeddings, sentence embeddings, and LLM-based embeddings, was compared. The results demonstrate that integrating LLMs surpasses all other models, enhancing the baseline version of the approach with a 45\% increase in F-measure.