基于大模型的高校图书馆个性化资源推荐系统构建与实践研究

打开文本图片集

Research on the Construction and Practice of University Library Personalized Resource Recommendation System Based on Large Models

NiuYue，Hao Baoquan， Zhao Zhiyan

AbstractIn the digitalage，universitylibraries are facing new challenges.The personalized resource recommendation system can acurately push information based onuser interests，habits，and needs，improving resourceutilization and experience.This paper proposes a recommendation system architecture based on large models，covering user profiling， resource indexing，and intelligentrecommendation.Thesystemutilizes temporalandmultimodalanalysis tocharacterize interests，and builds indexes withdeep understanding oflarge models tooptimize performance.In terms of deployment， explore cloud and localized applications，and demonstrate the effectiveness through caseanalysis.Compared to traditional methods，this system can accurately capture requirements and dynamically adjust strategies.The paper also discusses itsadvantages such as intellgenceand automation，analyzes risks such as privacy securityand content reliability，and looks forward tothedevelopmenttrendofpersonalizedrecommendationsystems insmart libraries， providing reference forthe application of artificial intelligence in the field of knowledge services.

KeywordsSmartlibray.Pesonalizedresourcerecommendation.Largelanguagemodel.Userprofiling.Resourcesemanticindexing

0 引言

在数字化和信息爆炸的时代，高校图书馆的职能已从传统的文献存储与借阅服务拓展为知识传播、学术研究支持及学习资源整合的智能化服务。（剩余11304字）