基于决策树算法的专利无效宣告风险特征识别  

Risk Feature Recognition of Patent Invalidation Based on Decision Tree Algorithm

作　　者：彭启宁 柳炳祥 付振康 冯广宇 PENG Qining;LIU Bingxiang;FU Zhenkang;FENG Guangyu(Intellectual Property Information Service Center,Jingdezhen Ceramic University,Jingdezhen 333001,China;School of Information Engineering,Jingdezhen Ceramic University,Jingdezhen 333403,China;School of Management and Economics,Jingdezhen Ceramic University,Jingdezhen 333403,China)

机构地区：[1]景德镇陶瓷大学知识产权信息服务中心,江西景德镇333001 [2]景德镇陶瓷大学信息工程学院,江西景德镇333403 [3]景德镇陶瓷大学管理与经济学院,江西景德镇333403

出　　处：《科学与管理》2023年第6期11-17,共7页Science and Management

基　　金：江西省研究生创新基金项目(YC2021-S532)。

摘　　要：“十四五”规划提出要着眼于抢占未来产业发展的先机,重点关注和培育先导性和支柱性产业。近几年,相关领域的专利申请数量不断增加,随之而来的专利侵权和专利无效宣告发生的案件数量也在不断增长。本文以侵权专利为切入点,构建专利无效宣告指标体系,建立基于机器学习中决策树算法的专利无效分类预测模型,对新兴产业中生物产业的专利侵权案件进行无效宣告分类预测。该模型在测试集上的平均F1值达到了0.9738,并得出了导致专利易发生无效宣告的特征指标的重要程度排序。The 14th Five-Year Plan(2021-2025)proposed to focus on seizing the opportunities for future industrial development,focusing on and nurturing leading and pillar industries.Therefore,in recent years,the number of patent applications in related fields has been increasing,and the number of cases of patent infringement and patent invalidation has also been increasing.This paper takes the patent infringement as the starting point,constructs the patent invalidation declaration index system,establishes the patent invalidation classification prediction model based on the decision tree algorithm in machine learning,and predicts the patent infringement cases of the biological industry in the emerging industry.The average F1 value of the model on the test set reaches 0.9738,and the importance ranking of the characteristic indicators that lead to patent invalidation is obtained.

关 键 词：专利 无效宣告 风险识别 预警体系 决策树 

分 类 号：G306[文化科学]