Rough Sets Clustering and Markov model for Web Access Prediction

Abstract

Discovering user access patterns from web access log is increasing the importance of information to build up adaptive web server according to the individual user’s behavior. The variety of user behaviors on accessing information also grows, which has a great impact on the network utilization. In this paper, we present a rough set clustering to cluster web transactions from web access logs and using Markov model for next access prediction. Using this approach, users can effectively mine web log records to discover and predict access patterns. We perform experiments using real web trace logs collected from www.dusit.ac.th servers. In order to improve its prediction ration, the model includes a rough sets scheme in which search similarity measure to compute the similarity between two sequences using upper approximation.