Predicting next page access by Markov models and association rules on web log data

Abstract

Mining user patterns of log file can provide significant and useful informative knowledge. A large amount of the research has been concentrated on trying to correctly predict the pages a user will request. This task requires the development of models that can predict a user’s next request to a web server. In this paper, we propose a method for constructing first-order and second-order Markov models of Web site based on past visitor behavior compare with association rules technique. This algorithm has been used to cluster Web site with similar transition behaviors and compares the transition matrix to an optimal size for efficient used to further improve the efficiency of prediction. From this comparison we propose a best overall method and empirically test the proposed model on real web logs.