Effective Personalization of web search based on Fuzzy Information Retrieval

Abstract — Information on the web is growing huge in size and retrieval of relevant information from the web is a big challenge for search engines. Search engines retrieve huge set of documents out of which very few are relevant due to small size input query having few keywords. User information needs are vague or imprecise and changes his query as users are not conscious of their exact needs of information and thus making it difficult to understand the information need of the user and hence the precision of search results is low. Fuzzy Logic is an ideal tool to manage imprecise and vague information. In this paper an approach is proposed using Fuzzy logic to Personalize the web Search based on clustered query sessions. The initial user input query is represented by fuzzy set and expanded with related words based on Fuzzy thesaurus. This Fuzzy expanded input query is used to select the cluster which is most similar to the information need of the user input query. The selected cluster is used to recommend the Fuzzy ranked set of documents. The user’s response to recommended search results is tracked to capture the user profile. The process of expansion of user profile based on fuzzy thesaurus and recommendations of Fuzzy ranked set of documents continue till the search is personalized to the information need of the user. Experiment was conducted on the data set of user query sessions captured in Academics, Entertainment and Sports domain and results confirm the improvement of precision of search results