Using Markov Chains for Link Prediction in Adaptive Web Sites

摘要

The large number of Web pages on many Web sites has raised navigational problems. Markov chains have recently been used to model user navigational behavior on the World Wide Web (WWW). In this paper, we propose a method for constructing a Markov model of a Web site based on past visitor behavior. We use the Markov model to make link predictions that assist new users to navigate the Web site. An algorithm for transition probability matrix compression has been used to cluster Web pages with similar transition behaviors and compress the transition matrix to an optimal size for efficient probability calculation in link prediction. A maximal forward path method is used to further improve the efficiency of link prediction. Link prediction has been implemented in an online system called ONE (Online Navigation Explorer) to assist users' navigation in the adaptive Web site.