Abstract: Technology evolution research can make the developing process, history and status of a specific field clear. It’s valuable for government and enterprises’ S&T strategy management. Patent citation analysis has some disadvantages such as it can’t judge the technology similarity of two patents accurately and the scope of analysis and richness of potential information is limit in technology evolution research. Text mining method can analyze text of patents deeply. It can make up the disadvantages of patent citation analysis to some degree. So we explore the combination of patent citation analysis and text mining method to research technology evolution. We construct C-T (Citation-Text) patent network based on patent citation matrix and patent text similarity matrix, then we do a cluster analysis and visualize the results. This paper aims to carry out the innovation of technology evolution research methods and enrich methodology of technology evolution research.

Key words: Technology Evolution, Patents Network, Patent Citation Analysis, Text Mining

摘要： 技术演进研究可用于梳理技术领域的发展脉络和内部技术活动的发展历史及现状，对政府和企业的科技战略管理具有重要意义。专利引文分析在技术演进研究中存在难以准确判断专利的技术主题相似度、分析的范围和潜在信息的丰富性有限等缺陷，而文本挖掘方法可以对专利的文本内容进行深度分析，能在一定程度上弥补专利引文分析的缺陷，因此探索将专利引文分析与文本挖掘方法相结合，在专利引用关系矩阵和专利文本相似度矩阵的基础上创建C-T（Citation-Text）专利网络，并对C-T专利网络进行聚类分析和可视化展示来研究技术的演进过程，旨在进行技术演进研究方法的创新，丰富技术演进研究的方法体系。

关键词: 技术演进, 专利网络, 专利引文分析, 文本挖掘