摘要
检索系统可以通过引入本体来弥补传统关键字检索语义匮乏的缺陷,然而,领域专家构建本体存在过程复杂、工期长、更新困难等弊端。为此,综合分析多种本体构建方法和技术,针对专利数据的特点给出一套半自动构建本体的方案,在此基础上提出基于半自动构建本体的专利信息检索系统的体系框架,描述系统原型的设计思想和检索流程,通过实验验证该系统能很好的扩充延伸检索词,明显地提高了检索效率以及查全率。
The ontology can be introduced into the retrieval system to make up the shortcoming of lack of semantic relationship in traditional keyword retrieval. However,nowadays the ontology is constructed by domain experts with limitations such as complex process,long in period, and difficulty to update. Therefore,in this paper it proposed a scheme of semi-automatic constructing ontology for patient information data after carrying the general analysis on several methods and technology of constructing ontology, and based on which, a framework of patent information retrieval system using this scheme was presented. This paper also describes the design ideas and retrieval process of the system prototype. In the end of the article, it reports some experiments verification that the system could extend the semantic meaning of keywords, and the retrieving efficiency and the recall rate has been improved distinctly.
出处
《计算机应用与软件》
CSCD
2009年第7期3-5,16,共4页
Computer Applications and Software
基金
国家自然科学基金项目(60673092)
