The information explosion on the Internet makes it hard for users to obtain required information from the web searched results in a more personalized way. For the same input word, most search engines return the same result to each user without taking into consideration user preference. For many users, it is no longer sufficient to get non-customized results.

It is crucial to analyze users? search and browsing behaviors based on searching keywords entered by users, the clicking rate of each link in the result and the time they spend on each site. To this end, we have proposed a method to derive user searching profiles. We have also proposed a mechanism to derive document profiles, based on similarity score of documents. In this paper, we discuss how to use our model to combine the user searching profiles and the document profile, with a view to presenting customized search results to the users.