知識圖譜構(gòu)建研究綜述

打開文本圖片集

中圖分類號：TP391 文獻標識碼：A 文章編號：2096-4706（2025）08-0117-10

Abstract：As a structured semantic knowledge base，the Knowledge Graph plays a key role in many fields such as informationretrval，intellgntquestionasweringandcommendationsystems.Thisapeviews tetheecorecopoents of KnowledgeGraphconstruction，informationextraction，knowledgefusion，andknowledgerasoning.Informationetraction technologyhasdevelopedfromrule-basedmethods toMachineLearing model，andthentoDepLeaingmodel.Itiscurently evolvingtowardsajoint EntityRelationshipExtractionmodel thatreduces erorpropagationandimprovesaccuracy.Inthepart ofknowledgefusion，thestrategiesofentitylinkingandkowledge mergingarediscussed，andtheproblemofentityrecogition is solved byentitydisambiguationand entity alignment.The sectionon knowledge reasoning analyzes the reasoning methods basedonrules，epresentationlearningandDeepLeaming，anditsaplcationinnewknowledge discoveryanderorinformation corection.Finallytehallengesinteonstuctionprocessaepontedout，andsuggestiosforutureesearchditiosare proposed to promote the development of knowledge graph research and application.

Keywords： Knowledge Graph; information extraction; knowledge fusion; knowledge reasoning; Deep Learning

0 引言

20世紀90年代，計算機網(wǎng)絡(luò)在世界各地得到普及，網(wǎng)絡(luò)信息資源日漸豐富，信息數(shù)據(jù)呈現(xiàn)規(guī)模海量、類型繁多和快速增長等特征。（剩余24060字）