基于语义关联树的分类查询扩展算法  

Classified Query Expansion Algorithm Based on Semantic Relation Tree

作　　者：任永功[1] 范丹[1] 武佳林[1] 

机构地区：[1]辽宁师范大学计算机与信息技术学院,大连116029

出　　处：《计算机科学》2009年第9期238-241,277,共5页Computer Science

基　　金：国家自然科学基金项目(60603047);辽宁省科技计划项目(2008216014);辽宁省教育厅高等学校科研基金(2008341);大连市优秀青年科技人才基金(2008J23JH026)资助

摘　　要：查询扩展技术中引入语义计算是一个重要的研究方向。针对现有解决方法普遍存在缺少主题知识、引入无关词以及筛选函数不恰当的问题,提出了一种结合主题选取与局部反馈方法的语义关联树模型,从语义的角度进行分类查询扩展。在传统方法基础上结合Web文本分类语料库进行了有主题的分类扩展,并改进了扩展词筛选函数,增加了阈值限定,有效控制了噪音。结合用户交互与局部反馈的方法不但减少了传统相关反馈中用户的工作量而且弥补了单纯局部反馈高度依赖于初次检索结果的缺陷。在SMART平台的实验结果表明,该方法相比一般的查询扩展算法查全率及查准率均有所提高。Introducing semantic computing technology into the query expansion is an important research direction. In this paper we presented a semantic relation tree model which combines with topic selection and local feedback method, classified expand query from the perspective of semantic. Traditional methods exist for the problems, such as the lack of knowledge in the topic, the introduction of irrelevant words and the filter functions are not proper. We introduced Web text classification into the semantic relation tree model to make subject expansion with improving the word filter function and increasing the threshold limit to control noise. The combination of user interaction with the local feedback method not only reduces the user＇s work in traditional relevance feedback method but also solves the problem of highly dependent primary retrieval result in local feedback. The experimental results on the SMART platform show that this method can increase the rate of recall and precision.

关 键 词：语义关联树 主题选取 查询扩展 WEB文本分类 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]