基于概念树的主题爬取技术研究  

Research on Focused Crawling Technology Based on the Concept Tree

作　　者：曾义聪[1] 杨贯中[1] 刘柯[1] 

机构地区：[1]湖南大学软件学院,长沙410082

出　　处：《科学技术与工程》2005年第12期785-790,796,共7页Science Technology and Engineering

基　　金：教育部科研基金(200146)资助

摘　　要：提出了一种新型主题爬取方法。采用概念树构造主题层次图,赋予待爬取URL对象以层次语义信息,按照语义相关性与重要性选择待爬取URL对象,搜索属于特定语义相关主题的重要Web文档的WWW子集。A new method of focused crawling is presented. A diagram of topic levels is built using the concept tree. URL objects to be crawled are endowed with semantic information about topic levels by the diagram and selected according to the semantic relevance and importance. It searches only the important subset of the WWW that pertains to a specific topic of semantic relevance

关 键 词：概念 概念树 主题爬取 主题层次 

分 类 号：TP393.03[自动化与计算机技术—计算机应用技术]