Complex Ontology Matching with Large Language Model Embeddings

2025-02-19

AI Summary

Semantic matching between ontologies and knowledge graphs suffers from insufficient expressivity. To address this, we propose an expressive semantic alignment method leveraging large language model (LLM) embeddings, the first to deeply integrate LLMs throughout the entire matching pipeline. Our approach reconstructs the architecture across three levels: label similarity, subgraph neighborhood matching, and entity-level alignment, while innovatively incorporating ABox-driven relation discovery to enhance semantic expressivity. Specifically, it synergistically combines LLM embeddings (e.g., from ChatGLM and Llama3), subgraph neighborhood encoding, multi-granularity embedding contrast (at token, sentence, and LLM levels), and ABox-guided relation mining. Evaluated on standard benchmarks, our method achieves a 45% improvement in F1 score over state-of-the-art baselines—substantially outperforming conventional word- and sentence-embedding approaches—and empirically validates the pivotal role of LLMs in high-expressivity semantic alignment.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

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.

Problem

• Enhancing ontology matching with LLMs
• Generating expressive correspondences efficiently
• Improving F-measure by 45% using LLMs

Innovation

• Integrates Large Language Models embeddings
• Enhances ontology matching with sub-graphs
• Improves F-measure by 45%