计算机工程与应用2026，Vol.62Issue(1)：47-67,21.DOI:10.3778/j.issn.1002-8331.2412-0384
知识图谱实体对齐研究综述:从传统方法到前沿技术
Review of Knowledge Graph Entity Alignment Research:from Traditional Methods to Cutting-Edge Technologies
摘要
Abstract
With the development of Internet and big data technologies,knowledge graphs,as a critical structured tool for describing entities and their relationships,have been widely applied across multiple domains.The entity alignment task in knowledge graphs aims to integrate entity information from heterogeneous knowledge graphs,addressing data silos and significantly enhancing the quality of knowledge graph construction and enabling cross-domain applications.This paper provides a comprehensive review of research progress in knowledge graph entity alignment.Firstly,it introduces the fundamental concepts and types of knowledge graphs.Subsequently,it elaborates on traditional entity alignment methods,including feature similarity-based approaches,machine learning-based techniques,and reasoning-based methodologies.The paper then focuses on entity alignment methods leveraging knowledge representation learning technologies,and explores challenges in aligning entities within multimodal and temporal knowledge graphs.Finally,it discusses the broad prospects of entity alignment in natural language processing and intelligent applications,as well as the potential of com-bining existing methods with emerging technologies to improve alignment accuracy and efficiency.关键词
知识图谱/实体对齐/自然语言处理/知识图谱融合Key words
knowledge graph/entity alignment/natural language processing/knowledge graph integration分类
信息技术与安全科学引用本文复制引用
丛烁,苏贵斌,柳林,王海龙..知识图谱实体对齐研究综述:从传统方法到前沿技术[J].计算机工程与应用,2026,62(1):47-67,21.基金项目
内蒙古自治区自然科学基金(2023LHMS06006,2024LHMS06015) （2023LHMS06006,2024LHMS06015）
基于机器学习的智能碳排放管理系统开发项目(20240043C). （20240043C）
