Personalized Web Search by User Interest Hierarchy

Most of the web search engines are designed to serve all users, independent of the needs of any individual user. Personalization of web search is to carry out retrieval for each user incorporating individual user’s interests. Current personalization techniques use the hierarchy of hyperlinks or map a user query to a set of categories etc. But, they do not use both the contents of a web page and a User Interest Hierarchy (UIH) implicitly. We propose a novel technique to order retrieved results depending on his/her interests, which use the contents of a web page and a UIH. The UIH is learnt implicitly from a bookmark. Experimental results determined that our approach to personalize web search satisfies user’s request more accurately than Google in retrieving both interesting web pages and potentially interesting web pages.