我国图书馆技术应用研究的革故鼎新——融合Sentence-BERT与动态主题模型的论文分析  

Innovative Evolution in the Application of Library Technologies in China:A Study Using Sentence-BERT and Dynamic Topic Models

作　　者：宋凯 冉从敬[2] 杜娟娟 Song Kai;Ran Congjing;Du Juanjuan(Shandong Normal University Library;School of Information Management,Wuhan University;Center for Scientific Research and Development in Higher Education Institutes)

机构地区：[1]山东师范大学图书馆,山东济南250358 [2]武汉大学信息管理学院,湖北武汉430072 [3]教育部高等学校科学研究发展中心

出　　处：《图书馆杂志》2025年第3期24-40,55,共18页Library Journal

基　　金：国家社会科学基金青年项目“基于知识元的高校专利质量智能判别及其推荐研究”(项目编号:23CTQ028)的研究成果之一。

摘　　要：为了全面、直观地刻画我国“图书馆技术应用”研究的进展与变迁,本文以CNKI中1998—2024年CSSCI论文为分析基础,首先,采用深度学习模型Sentence-BERT和动态主题模型(DTM),充分挖掘文本语义和主题特征。随后,运用K-means算法进行主题聚类,从宏观层面探讨“图书馆技术应用”研究的主题演化趋势。最后,绘制了主题词共现网络,从微观层面解析不同阶段的研究聚焦重点。结果表明,我国“图书馆技术应用”研究主要聚焦在9大主题,通过划分5个时间阶段可以发现,新兴技术的涌现对我国“图书馆技术应用”研究有显著推动作用,相关研究紧跟技术发展趋势,积极探索了新兴技术在图书馆中的应用场景,为国内图书馆提供更加高效、智能和个性化的管理与服务提供了借鉴。This paper provides a comprehensive and intuitive overview of the progress and evolution of library technology application research in China,based on an analysis of CSSCI-indexed papers from CNKI published between 1998 and 2024.The study employs Sentence-BERT and Dynamic Topic Models(DTM)to extract semantic and thematic features from the texts.Utilizing the K-means algorithm for topic clustering,it examines thematic evolution from a macro perspective.A co-occurrence network of key terms is subsequently constructed to analyze focal points at different stages from a micro perspective.The findings reveal that library technology application research in China centers on nine major themes,and identify five distinct phases of development.The study highlights the substantial influence of emerging technologies in driving advancements in the field.The research closely tracks technological developments,actively investigating new applications of these technologies in libraries,with the aim of enhancing efficiency,intelligence,and personalized services.

关 键 词：图书馆 技术应用 Sentence-BERT 动态主题模型 共现网络 

分 类 号：G353.1[文化科学—情报学] G250.7