Fuzzy Thesaurus In The The Vsm Text Information Retrieval Methods

As the main storage form, text information increases vastly. A number of tools and methods are developed to filter information junk out and retain the documents that users really want. Meanwhile, most keyword-based information retrieval methods always generate the large trash and miss much important information. To overcome the drawback resulting from the keyword-based IR model mentioned above, it is very valuable to develop new text information retrieval based on fuzzy synonym thesaurus.Referred to the current text IR, the paper introduces fuzzy synonym thesaurus to the text IR process. By summarizing the researches done by the foreign and domestic researchers, the paper chooses VSM and builds fuzzy synonym thesaurus. Then some modification has been made during the query vector in order to improve recall. In addition, fuzzy theory has been used to compute the word weighting. Finally, threshold is introduced to deal with the query result to meet the users' special need. In order to verify the effect of text IR based on fuzzy synonym thesaurus, a text information retrieval system has been designed. Through the method, we can retrieve relevant documents in a relatively narrow search space and meanwhile widen the coverage of the retrieval to the related documents that do not necessarily contain the same words as the query. By comparing the current text IR with new IR method developed in this paper, the conclusion has been made, that is, the retrieving results obtained from the new text IR method show some improvements in two metrics, precision and recall.