图书情报知识 ›› 2025, Vol. 42 ›› Issue (4): 102-112, 150.doi: 10.13366/j.dik.2025.04.102

• 学术聚焦(2)·生成式 AI 大模型结合外部工具和知识库的场景化应用 • 上一篇    下一篇

ChatGPT结合插件赋能信息分析的效果研究

冯海英1, 曹茹烨2, 曹树金1,2   

  1. 1.山东理工大学信息管理学院,淄博,255000;
    2.中山大学信息管理学院,广州,510006
  • 出版日期:2025-07-10 发布日期:2025-08-16
  • 通讯作者: 曹树金(ORCID: 0000-0003-1855-4522),博士,教授,研究方向:信息组织与信息检索,Email: caosj@mail.sysu.edu.cn。(
  • 作者简介:冯海英(ORCID: 0009-0007-1519-6746),硕士研究生,研究方向:信息组织与信息检索,Email: fhy06042023@163.com; 曹茹烨(ORCID: 0000-0002-7807-7652),博士研究生,研究方向:信息组织,Email: 421973288@qq.com。
  • 基金资助:
    本文系国家社会科学基金社科学术社团主题学术活动资助课题研究类项目“中国共产党历史知识图谱与知识索引构建研究”(21STA028)的研究成果之一。

A Study on the Effectiveness of Combining ChatGPT with Plugins for Information Analysis

FENG Haiying1, CAO Ruye2, CAO Shujin1,2   

  1. 1.School of Information Management, Shandong University of Technology, Zibo, 255000;
    2.School of Information Management, Sun Yat-Sen University, Guangzhou, 510006
  • Online:2025-07-10 Published:2025-08-16
  • Contact: Correspondence should be addressed to CAO Shujin, Email: caosj@mail.sysu.edu.cn, ORCID: 0000-0003-1855-4522
  • Supported by:
    This is an outcome of the project "Research on the Construction of Historical Knowledge Graph and Knowledge Index of the Communist Party of China"(21STA028)supported by National Social Science Foundation of China for the Thematic Academic Activity Funding Program for Social Science Academic Associations.

摘要: [目的/意义]分析使用ChatGPT及其插件赋能信息分析的效能,探索信息分析与新一代AI技术的融合发展,为信息分析技术方法的创新和发展提供参考。[研究设计/方法]以信息分析的工作领域为参考框架,设计了科学信息分析、专利信息分析、社会信息分析三类8项实验任务,由ChatGPT结合插件根据提示完成任务,同时采用情报学研究方法对其完成效果进行对比评估。[结论/发现] ChatGPT可以自动处理和分析大量的数据,提高信息分析的效率,但对于高度专业化或复杂的分析任务还存在不足。ChatGPT结合插件在复杂的信息分析任务中具有明显优势,可以处理ChatGPT自身无法做到或者是做得不完美的任务。[创新/价值]综合考察ChatGPT及其插件赋能信息分析的应用效果,从而为信息分析领域未来的发展方向提供启示。

关键词: ChatGPT, 生成式AI, 信息分析, 插件, 实验任务

Abstract: [Purpose/Significance] To analyze the effectiveness of using ChatGPT and its plugins to empower information analysis, this paper explores the integration and development of information analysis and the new-generation AI technologies, providing references for the innovation and development of information analysis technologies and methods. [Design/Methodology] With the work domains of information analysis as the reference framework, the study designed eight experimental tasks in three categories, namely scientific information analysis, patent information analysis, and social information analysis. These tasks were completed by ChatGPT combined with plugins according to the prompts. Meanwhile, the completion effectiveness was comparatively evaluated using the research methods of information science. [Findings/Conclusion] ChatGPT can automatically process and analyze large amounts of data to improve the efficiency of information analysis, but it is still insufficient for highly specialized or complex analysis tasks. ChatGPT combined with plugins has distinct advantages in complex information analysis tasks, being able to handle tasks that ChatGPT itself can't accomplish, or cannot accomplish perfectly. [Originality/Value] This study comprehensively examines the application effectiveness of combining ChatGPT with its plugins for information analysis, thereby providing insights for the future development direction of the information analysis field.

Keywords: ChatGPT, Generative AI, Information analysis, Plugins, Experimental task