Ranking web search results from personalized perspective

In this paper, we exploit the technique of data mining to mine frequent access patterns from user browsing behavior. In light of frequent access patterns, we develop a solution procedure to automatically extract user interests. Furthermore, in accordance with user interests mined and feedbacks of users, we propose a new algorithm with the idea of dynamically adjusting the ranking scores of Web pages. Specifically, algorithm PPR (standing for Personalized PageRank), is divided into four phases. The first phase assigns the initial weights based on user interests. In the second phase, the virtual links and hubs are created according to user interests. By observing user click streams, our proposed algorithm will incrementally reflect user favors for the personalized ranking in the third phase. To improve the accuracy of ranking, collaborative filtering is taken into consideration when the new query is submitted. By conducting simulation experiments, we have shown that algorithm PPR is not only very effective but also very adaptive in providing personalized ranking to users.