An Intelligent Meta-Search Engine with Personalization

The volume of documents in Internet are increasing every day . Again, each internet user has different search requirements. There are different approaches for creating user profiles by gathering user information through proxy servers or desktop robots . But both techniques require active participation of the user to install and configure these applications. In this work, we investigated the use of a less-expensive means of gathering user information for personalized or user specific search. Our study explored the effectiveness of personalized search based upon user profiles constructed from user search records. Individual user information such as queries submitted ,results returned and web pages selected from retrieved results were collected to create user profiles. These profiles were then used to re-rank the search results and the rank-order of the user-examined results before and after re-ranking were compared. Our study found that user profiles based on queries were as effective as those based on title and short descriptions. We also found that our personalized re-ranking resulted in an observable improvement in the rank-order of the user-selected results.