一种基于最大权匹配计算的信息检索方法

Method of information retrieval based on computing maximum-weight-matching

基于《知网》的词汇语义计算方法,提出了一种基于向量空间模型的文本信息检索新方法。方法的基本技术思想是通过计算关键词的语义相似度,并采用最大权匹配方法来计算查询向量和文本向量的相似度,作为相关文本的检索依据。该方法基于全局最优,使文本和查询向量中各词条的相似度总和最大,从而可以从整体上提高文本信息检索的准确率。论文还通过原型实验对该方法的有效性进行了验证。

Based on the computation of words-semantic similarity-of ＂How-net＂,a new method of document information retrievingbased on the model of vector space has been proposed.The basic ideas of the method are that,firstly we compute the similaritybetween keywords by the words-senmntic similarity of ＂How-net＂,and seeomtly compute the similarity between the query vectorand the document vector based on computing the nmxin^um-weight-matching.This algorithm is based on the optimization on over-all situation and it might find the maxinmm sum of each pair terms＇ similarity between the query vector and the document vec-tor.The experiment result has demonstrated the validity of the algorithm.