AN EFFECTIVE FUZZY WAY TO DOCUMENT RETRIEVAL

摘要

With the rapid development of the Internet, more and more information trash is engendered when people retrieve documents. In this paper, we try to establish an effective method to get some improvements by applying fuzzy theory to thesaurus construction. It stems from the normal keyword-based information retrieval method and adopts three new strategies. By surveying these three strategies, we find that building a synonym thesaurus is the most important task. In addition, we encounter another crucial problem in developing the document retrieval method. Very often there are several modifiers for a keyword in the given query, of which appropriate modifiers for the keyword should be recognized correctly. To solve this problem we ourselves develop and embed a new Parser module in the proposed document retrieval method. A simple experiment is conducted in terms of all strategies described above. The experiment result shows that the proposed fuzzy document retrieval method is comparatively effective.