基于大语言模型的知识图谱构建及在灾害应急中的应用

打开文本图片集

中图分类号：TP391 文献标识码：A 文章编号：2096-4706（2025）08-0146-07

Abstract：China isoneofthecountries withthe mostserious naturaldisasters inthe world.Typhoons，arthquakesfoods and ther disasters ocur frequently，posing a huge threat tothesafetyof people'slives and property.Facing thecomplexityand urgency ofdisasteremergencymanagement，Knowledge Graphtechnology has graduallybecomearesearch hotspot inthis field byvirtueofitsadvantagesiniformationintegration，elationsipingandknowledgereasonig.However，theeterogeity andqualityofisasterdataarenotufor，hichliitstheabilityoftraditioalNaturalLanguageProcessngmethodstotract Knowledge Graph information.Therefore，this paper proposes aLarge Language Model knowledge extraction method based on distributedpromptstrategy.Troughrequirementhints，domainknowledgehintsandfew-shothints，theLargeLanguageModel can automatically extract entitiesand relationships from unstructured data， which improves the automation and accuracy of KnowledgeGraphconstruction.InordertoverifytheefectofKnowledgeGraphbasedonLargeLanguageModelinpractical application，thispapertakes thetyphoondisasteremergencyplanasanexampletocostruct thecoresponding Knowledge Graph and its intelligent question and answer system.

Keywords：Knowledge Graph;Large Language Model; knowledge extraction; inteligent questionandanswer

0 引言

灾减灾要求。（剩余8910字）