图书情报知识 ›› 2026, Vol. 43 ›› Issue (4): 80-92.doi: 10.13366/j.dik.2026.04.080

• 图书、文献与交流 • 上一篇    下一篇

AI驱动的图书馆编目实践研究

宋琳琳, 谭卓熙, 李欣钰   

  1. 中山大学信息管理学院,广州,510006
  • 出版日期:2026-07-10 发布日期:2026-10-06
  • 通讯作者: 李欣钰(ORCID: 0009-0002-1588-4286),博士研究生,研究方向:AI应用与管理,Email: lixy825@mail2.sysu.edu.cn。
  • 作者简介:宋琳琳(ORCID: 0000-0003-4268-1927),博士,副教授,研究方向:信息组织与检索,Email: songlinl@mail.sysu.edu.cn; 谭卓熙(ORCID: 0009-0004-9926-2277),硕士研究生,研究方向:图书编目、数字人文,Email: tanzhx3@mail2.sysu.edu.cn。
  • 基金资助:
    本文系国家社会科学基金项目“图书馆编目数据化的理论与实践研究”(20BTQ017)的研究成果之一。

A Study of AI-Driven Library Cataloging Practices

SONG Linlin, TAN Zhuoxi, LI Xinyu   

  1. School of Information Management, Sun Yat-sen University, Guangzhou, 510006
  • Online:2026-07-10 Published:2026-10-06
  • Contact: Correspondence should be addressed to LI Xinyu, Email: lixy825@mail2.sysu.edu.cn, ORCID: 0009-0002-1588-4286
  • Supported by:
    This is an outcome of the project "Research on Theory and Practice of Library Cataloging Datafication"(20BTQ017)supported by National Social Science Foundation of China.

摘要: [目的/意义]探索人工智能技术在图书馆编目领域的应用实践,旨在优化编目流程、提高编目效率、提升编目质量。[研究设计/方法]在国内外相关研究和实践的基础上,从内容、标准、技术、流程、工具等多维度构建AI在图书馆编目任务中的应用框架,详细分析各场景中实践现状与发展态势。[结论/发现]在书目记录生成中选择合适的编目模型,可以自动创建元数据并优化编目流程;在元数据管理领域,命名实体识别、元数据增强、质量控制是其关键技术和方式;在自动化标引领域,开源工具和生成式AI的综合应用会有效增强传统标引工具的效能。遵循编目标准规范、推动关联数据与AI的融合、促进图书馆系统平台的转型升级、依赖编目人员的智慧支持是推进AI应用的基本要求。[创新/价值]结合编目数据化理论研究成果,提炼出编目工作中典型的AI应用场景,明确其关键领域与实施要点,为我国针对性地开展AI编目提供借鉴,也为宏观规划图书馆编目发展提供参考。

关键词: 图书馆编目, 元数据增强, 书目数据生成, 人工智能技术, 自动化标引

Abstract: [Purpose/Significance] This study explores the application of artificial intelligence technology in the field of library cataloging, with the aim of optimizing cataloging processes, improving cataloging efficiency and enhancing cataloging quality. [Design/Methodology] Building upon a review of relevant domestic and international research and practice, this study constructs a multi-dimensional framework for the application of AI in library cataloging tasks—covering content, standards, technology, processes and tools—and provides a detailed analysis of the current state of practice and future development trends within each application scenario. [Findings/Conclusion] In bibliographic record generation, selecting an appropriate cataloging model can automate the creation of metadata and optimize cataloging workflows; in metadata management, named entity recognition, metadata enrichment and quality control are key technologies and approaches; in automated indexing, the integrated application of open-source tools and generative AI will effectively enhance the performance of traditional indexing tools. It sets out the fundamental requirements for advancing its application: adhering to cataloging standards and specifications; promoting the integration of linked data and AI; facilitating the transformation and upgrading of library system platforms; and relying on the expert guidance of catalogers. [Originality/Value] Drawing on research findings in the field of library cataloging datafication, this study identifies typical AI application scenarios in cataloging work, clarifies the key areas and considerations of implementation, and provides guidance for the targeted development of AI-assisted cataloging in China, as well as serving as a reference for the macro-level planning of library cataloging development.

Keywords: Library cataloging, Metadata enrichment, Metadata generation, Artificial Intelligence, Automated indexing