基于专利功效矩阵的技术主题关联分析及核心专利挖掘  

Technical Topics Association Analysis and Core Patents Mining Based on Patent Technology-Effect Matrix

作　　者：许海云[1] 方曙[1] 

机构地区：[1]中国科学院成都文献情报中心

出　　处：《情报学报》2014年第2期158-166,共9页Journal of the China Society for Scientific and Technical Information

基　　金：中国科学院西部之光博士资助项目“基于内容分析的领域核心专利挖掘及专利权人竞合态势分析方法研究”(项目编号:Y4C0011001)研究成果之一

摘　　要：随着专利数量的快速增多，如何掌握技术主题间的关联和全貌，并快速有效的从海量专利文献中找出具有特定技术功效的核心专利簇具有重要意义。本文利用专利功效矩阵的技术维度和功效维度的关联规则分析获取领域内技术主题与达成功效主题的关联度，通过获取低支持度、高置信度的技术功效关联对弥补低共现一弱关系技术主题的遗漏。在此基础上，通过对技术功效主题组合的2模网络分析识别具有相同技术功效、相同技术或相同功效的核心专利或专利簇。最后以深海潜水器的水下导航系统为例做了实证分析，通过人工判读和领域专家的评价验证了本文所用方法可以帮助技术专家更有效的掌握技术主题关联，并找到具备特定技术功效的核心专利簇。With the patents rapid increasment, how to make clear the general view of technical topics, and to identify effectively the core patent clusters with specific technology-effect from massive patent literatures is of great significance. In this paper, we analyze the correlation degree to patents technical topics relationship by association rules, for this method could better support the low support but high confidence level co-occurrence topics. On this basis, through 2 mode network analysis to the combination of patent technology-effect theme, we can identify core patent clusters with the same effect or the same technology or the both at the same time. Finally, with the navigation system of deep-sea underwater submersible as an example, we make an empirical analysis, through manual interpretation and field experts evaluation to the empirical analysis results, we conclude that the proposed method in this article can effectively help technicians to master technical topics association and to get the core patent clusters with specific technology-effect.

关 键 词：主题挖掘 专利计量 功效矩阵 关联规则 核心专利 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]