世界科技研究与发展 ›› 2024, Vol. 46 ›› Issue (3): 386-408.doi: 10.16507/j.issn.1006-6055.2024.03.001

• 科技评价与评估 • 上一篇    下一篇

技术交叉主题特征识别研究进展

张娴1,2 李嘉晖1,2   

  1. 1.中国科学院成都文献情报中心,成都 610299;2.中国科学院大学经济与管理学院信息资源管理系,北京 100190
  • 出版日期:2024-06-25 发布日期:2024-07-03
  • 基金资助:
    国家社会科学基金“技术创新路径识别与预测的多元关系融合方法研究”(18BTQ067)

Research Progress on Identification Methods of Technology Crossing Topics

ZHANG Xian1,2   LI Jiahui1,2   

  1. 1. National Science Library ( Chengdu ) , Chinese Academy of Sciences, Chengdu 610299, China; 2. Department of Information Resources Management, School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China
  • Online:2024-06-25 Published:2024-07-03

摘要:

技术交叉主题识别研究对于技术创新研究意义重大。本文首先分析了技术交叉的内涵,随后对技术交叉主题识别主要方法进行了系统调研、归纳与分析,最后总结现有研究特点及优劣。研究发现,当前技术交叉主题识别主要方法有信息计量、引文分析、网络分析、文本挖掘、知识图谱、机器学习,涉及弱信号、突破性、颠覆性识别。主要特点包括:1)由形态特征层面的计量测度转向内容特征层面的文本主题挖掘;2)基于技术交叉特性的关联分析、文本挖掘方法等成为主流;3)多种方法与技术结合运用。技术交叉主题识别的未来研究方向包括:深入微观层面主题特征识别;多种方法的综合性创新运用;多元数据融汇;多领域异构大数据融合与应用关键技术。

关键词: 技术交叉, 主题识别, 技术融合, 网络分析, 文本挖掘

Abstract:

The Identification of technology-crossing topics is of great significance. This paper analyzes the content of technology-crossing firstly, then reviews the main methods of technology-crossing topic identification. Finally, it summarizes the advantages and disadvantages of existing research. The current main methods for technology-crossing topic identification include informetrics, citation analysis, network analysis, text mining, knowledge graph, and machine learning, which involve identifications of weak signals, breakthrough technology and disruptive technology. There are three main characteristics in existing research: 1) transition from morphological metrics to text content analysis; 2) association analysis and text mining methods becoming mainstream; 3) comprehensive application of multiple methods and technologies. Future research should strengthen the identification of micro-level thematic features, integrate multiple research methods innovatively, enhance the integration of multivariate data, and develop key technologies for the fusion and application of heterogeneous bigdata in multiple fields.

Key words: Technology Crossing, Topic Identification, Technology Integration, Network Analysis, Text Mining