Ontology-Enhanced Knowledge Graph Completion using Large Language Models

TL;DR

This work addresses incompleteness in knowledge graphs and the limitations of purely black-box LLMs for KG completion. It introduces OL-KGC, which combines neural structural embeddings with automatic ontological extraction to provide logic-guided reasoning within large language models. Key contributions include an automated ontology extraction pipeline, RDF-to-text mapping to supply symbolic knowledge, a linear adapter that injects structure into LLM prompts, and LoRA-based fine-tuning, all validated by extensive experiments showing state-of-the-art results. The approach advances neural-symbolic KG completion and provides reproducible benchmarks for future research.

Abstract

Large Language Models (LLMs) have been extensively adopted in Knowledge Graph Completion (KGC), showcasing significant research advancements. However, as black-box models driven by deep neural architectures, current LLM-based KGC methods rely on implicit knowledge representation with parallel propagation of erroneous knowledge, thereby hindering their ability to produce conclusive and decisive reasoning outcomes. We aim to integrate neural-perceptual structural information with ontological knowledge, leveraging the powerful capabilities of LLMs to achieve a deeper understanding of the intrinsic logic of the knowledge. We propose an ontology enhanced KGC method using LLMs -- OL-KGC. It first leverages neural perceptual mechanisms to effectively embed structural information into the textual space, and then uses an automated extraction algorithm to retrieve ontological knowledge from the knowledge graphs (KGs) that needs to be completed, which is further transformed into a textual format comprehensible to LLMs for providing logic guidance. We conducted extensive experiments on three widely-used benchmarks -- FB15K-237, UMLS and WN18RR. The experimental results demonstrate that OL-KGC significantly outperforms existing mainstream KGC methods across multiple evaluation metrics, achieving state-of-the-art performance.

Figures (5)

• Figure 1: LLM-based methods that fail to effectively utilize ontological knowledge.
• Figure 2: The role of ontological knowledge in reasoning.
• Figure 3: The overall framework of OL-KGC.
• Figure 4: The results of the ontology rule ablation experiments.
• Figure 5: The results of the ontological knowledge analysis experiment.