基于概念类的索引方法  

Information Retrieval Based on Concept Cluster

作　　者：孙铁利[1] 曹晶[1] 

机构地区：[1]东北师范大学计算机学院,长春130117

出　　处：《长春理工大学学报（自然科学版）》2006年第1期79-83,共5页Journal of Changchun University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(69973012)

摘　　要：传统的信息检索方法一般都采用对文本内容的词频进行分析的统计方法,这种索引方法仅仅考虑词语在文本中的出现率,因此不能抽取出表达文本语义的索引词。为了解决这个问题,本文提出了一种新的信息检索方法,即基于概念的权重索引方法。本方法引入了概念类的概念,并且提出了用概念之间存在的关系来表示文档中的词汇和概念的语义重要度。本方法比单纯的词汇信息更能体现文本的概念特征,提高信息检索的性能;同时还能降低文本向量的维数,减少计算量,提高检索效率。Traditional approaches to index weighting for information retrieval from texts are based on statistical of the texts＇ contents, A key shortcoming of these indexing schemes,which consider only the terms in a document,is that they cannot extract semantically indexs that represent the semantic content of a document. To address this issue, we proposed a new indexing formalism that considers not only the terms in a document, but also the concepts. In the proposed method, concept are extracted by exploiting clusters of terms that are semantically related, reffered to as concept clusters, and deal the link information of a concept cluster as means to measure the importance degree of a word. Therefore, the proposed method can improve the performance of the information retrieval,and reduce the dimension of index terms.

关 键 词：索引权重 词汇链 概念类 检索 WORDNET 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]