To improve the performance of concept retrieval in search engine, a new algorithm construction of synonym identification is present in this paper.Keyword retrieval is the most popular method in search engine retrieval, but most search engines do not controll vocabulary in information retrieval. This makes the retrieval efficiency not competent. The resolution of this problem is to use "concept retrieval" instead of "Word-form retrieval"The new algorithm construction of synonym recongnition can makes computer system recognize synonyms automaticly. To do this, a collection of word units was established. A word unit based semantic dictionary is also required. In the system, compound words(phrase) are cuted to word units, and transformed to semantics codes in order to do semantic comparison. Each word units has its own calculating weight by the position and the function they stand in the whole phrase.To judge the effect of the new algorithm, a comparing experiment was designed. The results show that the algorithm is competent. And it抯 significant to use in intelligent search engine.
|