基于权利要求书的专利快速自动分类方法  

Fast Automatic Classification Method of Patents Based on Claims

作　　者：徐丹[1] XU Dan

机构地区：[1]中国国家知识产权局专利局光电部,北京100086

出　　处：《图书情报导刊》2018年第10期68-72,共5页Journal of Library and Information Science

摘　　要：随着国家对知识产权的重视,专利申请量呈高速增长的趋势,对专利快速分类的需求越来越迫切。基于权利要求书的专利快速分类方法,充分考虑了专利特性及内容细节,结合专利分类层次结构特性,建立了特征词汇表及其权重矩阵,根据专利文本特征词汇分布情况,并结合权重矩阵进行分类。实验结果表明:该方法在专利不同层次上的分类效果显著,且能够满足专利快速分类的实时性要求。With the country's emphasis on intellectual property rights,the amount of patent applications shows a trend of rapid growth,and the demand for rapid classification of the patents is becoming increasingly urgent.The fast classification method based on patent claims,takes full account of patent characteristics and content details,establishes the feature vocabulary and its weight matrix combining the hierarchical structure of patent classification,and classifies the patent text according to the distribution of feature vocabulary and the weight matrix.The experimental results show that the method has a significant classification effect at different levels of patents and can meet the real-time requirements of patent fast classification.

关 键 词：权利要求书 特征词 自动分类 权重矩阵 

分 类 号：G306[文化科学]