计算机应用研究2025，Vol.42Issue(10)：3005-3011,7.DOI:10.19734/j.issn.1001-3695.2024.11.0537
融合实体特征聚合和关系语义聚合的推理模型
Reasoning model integrating entity feature aggregation and relation semantic aggregation
摘要
Abstract
Most existing temporal knowledge graph reasoning models rely on relational graph neural networks to capture se-mantic dependencies between entities in each snapshot.To better utilize structural information within graph data,this paper proposed the EFRSA reasoning model,which integrated entity feature aggregation and relational semantic aggregation.This model effectively captured semantic dependencies among concurrent entities at each timestamp.Through its entity feature ag-gregation module,EFRSA identified and leveraged the potential significant associations among co-occurring entities.Additio-nally,EFRSA introduced a relation semantic aggregation module based on relational subgraph associations to fully express rela-tional semantic information in the graph structure.Experimental results on datasets such as ICEWS14,GDELT,YAGO,and WIKI show that EFRSA achieves an MRR improvement of 0.89～3.24 in entity prediction and outperforms other methods in relation semantic prediction,thereby enhancing the model's reasoning capability.关键词
时序知识图谱/图结构/实体特征聚合/关系语义聚合Key words
temporal knowledge graph/graph structure/entity feature aggregation/relation semantic aggregation分类
计算机与自动化引用本文复制引用
董文永,梁智学,周孟强,贾亚洁..融合实体特征聚合和关系语义聚合的推理模型[J].计算机应用研究,2025,42(10):3005-3011,7.基金项目
国家自然基金面上项目(61672024) （61672024）
国家重点专项研发计划资助项目(2018YFB2100500) （2018YFB2100500）
