情报杂志2025，Vol.44Issue(5)：139-146,138,9.DOI:10.3969/j.issn.1002-1965.2025.05.017
基于技术融合视角的颠覆性专利预测研究
Research on Disruptive Patent Prediction Based on Technology Convergence Perspective
摘要
Abstract
[Research purpose]Technology convergence is an important source of disruptive technologies.Scientific and accurate predic-tion of disruptive technologies spawned by technological integration is of great significance for China to accelerate scientific and technologi-cal integration and progress in specific technological fields and achieve"lane overtaking".[Research method]This paper constructs a disruptive patent prediction framework from the perspective of technology convergence.Firstly,based on the Rao-Stirling index,the pa-tent technology fusion intensity is measured.Secondly,the core features of the technology fusion subversive technology are described by combining literature metrology and text mining methods.Indicators are constructed to identify historical disruptive patents from three di-mensions:novelty,mutability and high impact.Finally,the Autogluon machine learning framework is used to fit the potential relationship between the external features of patent applications and disruptive patents,and to complete the prediction of disruptive patents.[Research rusult/conclusion]An empirical research was carried out in the field of quantum information,and a total of 377 disruptive patents are pre-dicted,covering 7 types of technologies such as quantum computing processors,single photon detection,and quantum dot materials,pro-viding a reference for decision makers to grasp the direction of technology integration and promote the innovation and development of dis-ruptive technologies.关键词
技术融合/颠覆性专利/量子信息/Rao-Stirling指数/Autogluon/机器学习Key words
technology integration/disruptive patent/quantum information/Rao-Stirling index/Autogluon/machine learning引用本文复制引用
方曦,彭康,刘云..基于技术融合视角的颠覆性专利预测研究[J].情报杂志,2025,44(5):139-146,138,9.基金项目
国家自然科学基金重点国际(地区)合作研究项目"新兴产业全球创新网络形成机制、演进特征及对创新绩效的影响研究"(编号:71810107004) （地区）
科技部创新方法工作专项"科技成果价值评估方法与应用示范研究"(编号:2020IM021000)研究成果. （编号:2020IM021000）
