Recommendation of optimized web pages to users using Web Log mining techniques

Now a days, user rely on the web for information, but the currently available search engines often gives a long list of results, much of which are not always relevant to the user's requirement. Web Logs are important information repositories, which record user activities on the search results. The mining of these logs can improve the performance of search engines, since a user has a specific goal when searching for information. Optimized search may provide the results that accurately satisfy user's specific goal for the search. In this paper, we propose a web recommendation approach which is based on learning from web logs and recommends user a list of pages which are relevant to him by comparing with user's historic pattern. Finally, search result list is optimized by re-ranking the result pages. The proposed system proves to be efficient as the pages desired by the user, are on the top in the result list and thus reducing the search time.