A Semantic based query expansion to search

Keyword based information retrieval has difficulties in retrieving relevant information because it is not able to include the semantics of queries. In this paper, we present a novel method for query expansion based on semantic relations. In our proposed algorithm, semantically related words to the query are extracted from WordNet. Valuable words among extracted words are selected as candidate expansion terms. At last candidate terms which do not cause ambiguity and noise in the query are selected as expansion words. This approach is naturally robust to noise words and can improve semantic inferring of information retrieval.