Enhancing Knowledge Graph Completion with Entity Neighborhood and Relation Context
Jianfang Chen, Kai Zhang, Aoran Gan, Shiwei Tong, Shuanghong Shen, Qi Liu
TL;DR
Knowledge Graph Completion remains challenged by incompleteness and scalability. The paper presents KGC-ERC, a sequence-to-sequence framework that injects both entity neighborhood and relation context into the input of a generative language model, guided by a Selector-based sampling strategy to fit within input token limits. It introduces tailored sampling for entity neighbors and relation contexts, substantively improving reasoning and predictive accuracy across Wikidata5M, Wiki27K, and FB15K-237-N while maintaining a compact 60M parameter footprint. The experimental results demonstrate strong performance gains and scalability, with ablation studies confirming the importance of relation context and the sampling strategy for maximizing contextual utility.
Abstract
Knowledge Graph Completion (KGC) aims to infer missing information in Knowledge Graphs (KGs) to address their inherent incompleteness. Traditional structure-based KGC methods, while effective, face significant computational demands and scalability challenges due to the need for dense embedding learning and scoring all entities in the KG for each prediction. Recent text-based approaches using language models like T5 and BERT have mitigated these issues by converting KG triples into text for reasoning. However, they often fail to fully utilize contextual information, focusing mainly on the neighborhood of the entity and neglecting the context of the relation. To address this issue, we propose KGC-ERC, a framework that integrates both types of context to enrich the input of generative language models and enhance their reasoning capabilities. Additionally, we introduce a sampling strategy to effectively select relevant context within input token constraints, which optimizes the utilization of contextual information and potentially improves model performance. Experiments on the Wikidata5M, Wiki27K, and FB15K-237-N datasets show that KGC-ERC outperforms or matches state-of-the-art baselines in predictive performance and scalability.