Angielski
Internet helps us to get all the required information; but the question is that whether the retrieved information is relevant or not. Query recommendation system helps user in finding the accurate information from the Web corpus efficiently. Search engine is used when user know their actual query. User prefers to use recommendation system when they don't know what query must be fired to search engine. Thus query recommendation assists the learner to satisfy the information need at very initial stage. Many researchers have presented different methods for recommendation based on history of user, navigational patters, query flow graph, query similarity, user previous query, click behaviors, history of snippets for information retrieval. But sometimes these methods fail to retrieve the proper recommendation for low frequency queries. The presented method for URL recommendation uses snippets, user preference, location and the synonyms to recommend better results. The performances of the systems are measured on the basis of precision and recall. The method works better and effective for all high and low frequency queries.
