基于知识图谱和RAG技术的古诗词智能问答系统

打开文本图片集

中图分类号：TP391.1 文献标识码：A 文章编号：2096-4706（2025）20-0115-07

Abstract：In viewofthe hallcination problems such asauthor confusionanddynastydislocation in thefieldof ancient poetry，this project innovatively proposes a solution based on Retrieval-Augmented Generation （RAG） technology. By constructing a professional Knowledge Graph ofancientpoetryand designing multi-dimensional prompt engineering，thesystem caneffectivelycorrect theerrorsoftheLargeLanguage Model inthecomplexsemanticunderstandingof poetrybackground， historicalcotextadspcts.InsofseificiplemetatiofstlymedEntityecogiioisuedttctt keyinformationofthe problem，andthenauthoritativecontentisobtained through KnowledgeGraphretrievalasthebasis for generation.Finall，RAGas evaluationverifiesthatthehallucinationrateof theLargeLanguage Modelissignificantlyreduced. This studyflls thegapintheacuracyof theapplicationoftheLargeLanguageModelinthefeldof traditional cultureand provides areliable technical path forAI to enable cultural inheritance.

Keywords： hallcination problem; Retrieval-Augmented Generation; Natural Language Processing; Knowledge Graph Large Language Model; Named Entity Recognition

0 引言

随着大语言模型在自然语言处理领域的迅猛发展，其在各类智能问答系统中的表现已获得广泛关注，且智力水平呈现“涌现”现象[。（剩余11031字）