一种基于同义词词典的模糊查询扩展方法

An approach to fuzzy query expansion based on synonymy thesaurus

在信息检索系统中,查询扩展是一种非常有效的改进检索性能的方法.为此,提出一种基于同义词词典的模糊查询扩展方法.该方法中的同义词词典是基于著名的语义词典WordNet中的同义词集合建立的,同义词之间的贴近度[0,1]使用Tanimoto系数获得.利用该词典,能够进行较好的查询扩展.将该方法与向量空间模型结合应用于文本信息检索系统中,所构造的检索模型相当于一种简单的语义模型,并且可以根据阈值来控制查询扩展的程度.所得试验结果表明,使用该查询扩展方法的信息检索系统较常规信息检索系统的检索性能有一定改善.

Query expansion （QE） has been proved to be one of effective methods for improving the performance of the information retrieval （IR） system. Therefore, a new fuzzy QE method based on synonymy thesaurus is proposed, and the synonymy thesaurus is built based on the famous lexical database WordNet. In the synonymy thesaurus, the similarity between the synonyms is [0, 1], which is obtained by Tanimoto coefficient. By using this synonymy thesaurus, query expansion can be done well. Then the fuzzy QE method is introduced into the document information retrieval system together with the modified vector space model. The experimental results show that the developed information retrieval system has got more effective performance than before by using the fuzzy query expansion method. One feature of the proposed information retrieval model is that it can be treated as one of simple semantic models. Another feature is that the expansion degree is controllable based on different thresholds.