基于专利本体的语义检索研究

A Research on Semantic Retrieval Based on Patent Ontology

国家知识产权局CNIPR等著名专利检索系统仍是基于关键词的传统检索方式,会漏掉很多重要的相关专利数据。为提高专利信息的利用率,文章提出一种基于专利本体的语义检索模型,并采用Java语言编程实现,最后设计实验测试其检索效果。实验结果表明本研究能根据专利本体库中已经存在的知识和人为定义的规则,推导出库中原来没有的新的隐含知识,有效提高了相关信息的查全率和查准率。

The CNIPR patent retrieval system of the State Intellectual Property Office is still based on traditional keywords retrievat methods, which will miss a lot of important related patents. To improve the utilization of patent information, this paper makes a research on semantic retrieval based on patent ontology and realizes the semantic retrieval system by programming language Java. Finally, we design experiments to verify the application effect of the semantic retrieval system. The experimental results show that compared with the traditional way of single retrieval based on keywords, this study can effectively improve the recall ratio and precision ratio of patent information, and it can retrieve out the new implicit knowledge according to existing knowledge and human defined rules in the patent ontology.