摘要
[目的/意义]专利作为知识产权的核心内容,是全球技术、经济和战略竞争的重要基础资源。近年来,专利数量激增,专利信息的“科技金矿”地位持续受到质疑,这也使得海量专利数据中的知识发现变得越来越有价值。本文对国际高价值专利研究进行全面深入解读,有利于为研究者、企业、技术人员提供参考。[方法/过程]本文首先综合使用TF-IDF算法和LDA主题模型挖掘国际高价值专利研究的热点主题,然后基于扎根理论对国际高价值专利研究热点主题及文献进行自下而上的三级编码,实现对国际高价值专利研究主题范畴的归纳与综述。[结果/结论]国际高价值专利研究主题形成了7个主题范畴,其中专利价值评估和核心专利价值预测两大类主题为国际高价值专利研究的热点主题范畴;专利价值评估包括专利多维价值评价指标、专利价值评估方法、专利质量影响因素3个方面;核心专利价值预测包括核心专利特征研究以及核心专利识别方法,主要识别方法有机器学习方法、文本挖掘方法和引文网络分析等。
[Purpose/Significance]As the core content of intellectual property rights,patents are an important fundamental resource for global technological,economic,and strategic competition.In recent years,the number of patents has skyrocketed,and the status of patent information as a“technology gold mine”continues to be questioned,making knowledge discovery in massive patent data increasingly valuable.The article provides a comprehensive and in-depth interpretation of international high-value patent research,which is beneficial for providing reference for researchers,enterprises,and technical personnel.[Method/Process]This paper first comprehensively used TF-IDF algorithm and LDA topic model to mine hot topics of international high-value patent research,and then used the Grounded theory to carry out bottom-up three-level coding of international high-value patent research hot topic and literature to achieve an overview of international high-value patent main research topic categories.[Result/Conclusion]International high-value patents have formed 7 thematic categories,with two main categories of topics:patent value evaluation and core patent value prediction being hot topics in international high-value patent research;Patent value evaluation includes 3 aspects:multi-dimensional evaluation indicators of patent value,patent value evaluation methods,and factors affecting patent quality;The prediction of core patent value includes research on core patent features and core patent recognition methods,mainly including machine learning methods,text mining methods,and citation network analysis.
作者
陈玉
胡泽文
周西姬
Chen Yu;Hu Zewen;Zhou Xiji(School of Management Science and Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,China)
出处
《现代情报》
CSSCI
北大核心
2024年第8期153-170,共18页
Journal of Modern Information
基金
国家社会科学基金项目“面向海量科技文献的潜在‘精品’识别方法与应用研究”(项目编号:20CTQ031)
江苏省教育厅江苏高校“青蓝工程”项目资助
江苏省高校哲学社会科学研究重大项目“江苏省未来产业高价值专利智能识别与培育机制研究”。
