Web search personalization using machine learning techniques

Information on the web is increasing at an enormous speed. Every user has a distinct background and aspecific goal when searching for information on the web. Present search engines produce results that are best suited to given query. But these engines are unaware of user´s individual preferences which in turn can vary with individual interest and these interests most of the time change with individual working environment time. To provide such personalized results, user´s topical preferences could be stored and utilized for the purpose. Different approaches have been implemented for the same such as, Collaborative Filtering, Document-Based or Concept based profiling etc. We are proposing hybrid approach based on Document Based as well as Concept Based Profiling. Proposed framework aims to re-rank results for a given query obtained from existing search engines. Thus this system would provide an adaptive methodology for learning changing user preferences to re-rank results according to one´s individual interests.