Abstract: The way of getting different and explosive growth of Web information effectively is more critical. The first challenge is that it is not easy to recommend latent semantically relevant results to users. In the Query Suggestion, there are several outstanding issues that can potentially degrade the quality of the recommendations. The first one is the ambiguity which commonly exists in the natural language. Queries containing ambiguous terms may confuse the algorithms which do not satisfy the information needs of users. This research cope with the abbreviation based Query suggestion to provide the recommendation for user query, based on the short term query word. It also proposes K-means clustering Algorithm for generating user group based query matrix. In previous search query words are also taken into query suggestion calculation. In this previous query based suggestion method not consider the Zero Visit count URLs. But in normal find query suggestion also consider the Zero visit count URLs. The Graph query suggestion construct the Bipartite directed graph, based on the given queries and the access URLs. In addition Find Query Suggestion is constructed for domain wise. In addition, to add the recommendation user details play an important role in online query suggestion calculation. The view recommendation user URL access details to provide the user URL access details, it is used to the user can easily identify which one is the best URL for the given query. At run time user URL access matrix to provide all users URL access details and user group based query matrix which is used to give the user access details by group wise. In addition the image can be searched based on the suggested query words. In this proposed method the images can be added under the category wise and image search results returns all the images which is added under the category.


Keywords: visit count, query suggestion, URL, image, short term query word.