基于专利合成引文的技术聚类分析——以人工智能领域为例

doi:10.3772/j.issn.1000-0135.2026.08.003

情报学报

2026, Vol. 45

Issue (8): 1106-1125 DOI: 10.3772/j.issn.1000-0135.2026.08.003

情报理论与方法

本期目录 | 过刊浏览 | 高级检索

基于专利合成引文的技术聚类分析——以人工智能领域为例

甘静娴¹, 田琛², 王昱³

1.大连理工大学知识产权学院，大连 116024
2.大连工业大学管理学院，大连 116034
3.大连理工大学商学院，盘锦 124221

Technical Clustering Based on Patent Synthetic Citations: Case Study in Artificial Intelligence

Gan Jingxian¹, Tian Chen², Wang Yu³

1.School of Intellectual Property, Dalian University of Technology, Dalian 116024
2.School of Management, Dalian Polytechnic University, Dalian 116034
3.School of Business, Dalian University of Technology, Panjin 124221

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (6727 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对现有专利技术主题聚类中真实引文数据存在的覆盖稀疏与技术相关性表征不足等局限，本文提出一种基于全文本相似度的合成引文构建算法，并进一步结合文献计量学中的耦合逻辑，开发了合成引文双向耦合聚类模型。该方法旨在通过挖掘潜在的语义关联，在虚拟语义空间中重构专利间的技术引用网络。研究发现，相较于传统真实引文网络，合成引文网络不仅能有效捕捉被传统方法遗漏的潜在技术关联，更能显著提升专利间技术相关性的测度精度。实证结果表明，基于该方法的聚类结果在主题识别的完整性与分类准确性上均实现了质的提升。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	甘静娴
	田琛
	王昱

关键词 ：合成引文, 专利耦合, 技术聚类, 人工智能, 文本分析

收稿日期: 2025-09-25

基金资助:国家自然科学基金项目“动态一致性视角的企业高价值专利测度及内外部生成机制研究”（72202026）；教育部人文社会科学研究青年基金项目“新兴技术背景下技术标准化驱动的企业创新韧性研究：测度评估、因果效应和协同机制”（25YJCZH235）；辽宁省经济社会发展研究课题青年课题“大力培育新质生产力，推动辽宁战略性新兴产业、未来产业布局与发展对策研究”（2025lslqnkt-057）。

作者简介: 甘静娴，1988年生，博士，讲师，硕士生导师，主要从事专利挖掘、政策分析和大数据分析研究；田琛，通信作者，1991年生，博士，讲师，硕士生导师，主要从事文本挖掘和技术标准化研究，E-mail：tianchen@dlpu.edu.cn；王昱，1984年生，博士，副教授，硕士生导师，主要从事数字经济和技术创新研究；

引用本文:

甘静娴, 田琛, 王昱. 基于专利合成引文的技术聚类分析——以人工智能领域为例[J]. 情报学报, 2026, 45(8): 1106-1125.
Gan Jingxian, Tian Chen, Wang Yu. Technical Clustering Based on Patent Synthetic Citations: Case Study in Artificial Intelligence. 情报学报, 2026, 45(8): 1106-1125.

链接本文:

https://qbxb.istic.ac.cn/CN/10.3772/j.issn.1000-0135.2026.08.003 或 https://qbxb.istic.ac.cn/CN/Y2026/V45/I8/1106

1 韦婷婷, 冯丹钰, 宋世领, 等. 技术机会发现领域专利挖掘方法研究述评[J]. 情报学报, 2023, 42(10): 1238-1250.
2 赵又霖, 顾陈娅, 王嘉杰, 等. 技术融合机会识别研究: 关系嵌入和结构嵌入的综合考量[J]. 情报学报, 2025, 44(10): 1315-1328.
3 Fleming L. Recombinant uncertainty in technological search[J]. Management Science, 2001, 47(1): 117-132.
4 Kang I S, Na S H, Kim J, et al. Cluster-based patent retrieval[J]. Information Processing & Management, 2007, 43(5): 1173-1182.
5 高楠, 傅俊英, 赵蕴华. 融合专利共被引和耦合方法的研究前沿识别——以脑机接口领域为例[J]. 情报学报, 2016, 35(9): 971-979.
6 谢士尧, 王小梅. 基于深度文本聚类的论文与专利数据融合方法研究[J]. 数据分析与知识发现, 2024, 8(4): 112-124.
7 Kelly B, Papanikolaou D, Seru A, et al. Measuring technological innovation over the long run[J]. American Economic Review: Insights, 2021, 3(3): 303-320.
8 逯万辉. 基于专利引文网络挖掘的技术研发路径识别与颠覆性创新信号探测研究[J]. 情报学报, 2024, 43(9): 1059-1069.
9 Garfield E. Citation indexes for science: a new dimension in documentation through association of ideas[J]. Science, 1955, 122(3159): 108-111.
10 李昊, 侯剑华, 张洋. 引文中辍的理论基础与实证研究[J]. 情报学报, 2025, 44(6): 702-719.
11 Gan J X, Zhang Y J. Dynamic analysis of patent value evolution: a study on the roles of novelty and impact[J]. Technology Analysis & Strategic Management, 2026, 38(7): 1070-1101.
12 Cotropia C A, Lemley M A, Sampat B. Do applicant patent citations matter?[J]. Research Policy, 2013, 42(4): 844-854.
13 Randi? M. Citations versus limitations of citations: beyond Hirsch index[J]. Scientometrics, 2009, 80(3): 809-818.
14 Robinson-Garcia N, Mongeon P, Jeng W, et al. DataCite as a novel bibliometric source: coverage, strengths and limitations[J]. Journal of Informetrics, 2017, 11(3): 841-854.
15 Martínez M A, Herrera M, López-Gijón J, et al. H-classics: characterizing the concept of citation classics through H-index[J]. Scientometrics, 2014, 98(3): 1971-1983.
16 Zhang Y, Huang Y, Porter A L, et al. Discovering and forecasting interactions in big data research: a learning-enhanced bibliometric study[J]. Technological Forecasting and Social Change, 2019, 146: 795-807.
17 Suominen A. Phases of growth in a green tech research network: a bibliometric evaluation of fuel cell technology from 1991 to 2010[J]. Scientometrics, 2014, 100(1): 51-72.
18 Ponomariov B, Boardman C. What is co-authorship?[J]. Scientometrics, 2016, 109(3): 1939-1963.
19 Boyack K W, Klavans R. Co-citation analysis, bibliographic coupling, and direct citation: which citation approach represents the research front most accurately?[J]. Journal of the American Society for Information Science and Technology, 2010, 61(12): 2389-2404.
20 Weinberg B H. Bibliographic coupling: a review[J]. Information Storage and Retrieval, 1974, 10(5/6): 189-196.
21 Jarneving B. Bibliographic coupling and its application to research-front and other core documents[J]. Journal of Informetrics, 2007, 1(4): 287-307.
22 Gan J X, Qi Y. Selection of the optimal number of topics for LDA topic model—taking patent policy analysis as an example[J]. Entropy, 2021, 23(10): 1301.
23 Furukawa T, Mori K, Arino K, et al. Identifying the evolutionary process of emerging technologies: a chronological network analysis of World Wide Web conference sessions[J]. Technological Forecasting and Social Change, 2015, 91: 280-294.
24 Shibata N, Kajikawa Y, Takeda Y, et al. Comparative study on methods of detecting research fronts using different types of citation[J]. Journal of the American Society for Information Science and Technology, 2009, 60(3): 571-580.
25 Huang M H, Chang C P. A comparative study on detecting research fronts in the organic light-emitting diode (OLED) field using bibliographic coupling and co-citation[J]. Scientometrics, 2015, 102(3): 2041-2057.
26 Hu X J, Rousseau R, Chen J. On the definition of forward and backward citation generations[J]. Journal of Informetrics, 2011, 5(1): 27-36.
27 Zhang Y, Wang Y, Du H F, et al. Delayed citation impact of interdisciplinary research[J]. Journal of Informetrics, 2024, 18(1): 101468.
28 甘静娴. 专利政策、研发行为对企业专利质量的影响机理研究[D]. 南京: 南京理工大学, 2021.
29 Zhou K Z, Li C B. How knowledge affects radical innovation: knowledge base, market knowledge acquisition, and internal knowledge sharing[J]. Strategic Management Journal, 2012, 33(9): 1090-1102.
30 Giczy A V, Pairolero N A, Toole A A. Identifying artificial intelligence (AI) invention: a novel AI patent dataset[J]. The Journal of Technology Transfer, 2022, 47(2): 476-505.