Integrating automatic Web page clustering into Web log association mining

An immense number of user accesses to WWW resources as a result of the growth of the internet has led to the great need of finding useful patterns and rules of user behaviors, and resulted in the rise of technologies for Web usage mining. The goal of Web usage mining is to discover the useful information of Web surfer's sessions and behaviors from the Web server transaction records. Current Web usage mining applications rely exclusively on log files. My hypothesis is that Web page contents can be used to improve Web usage mining results. I proposed a system that integrates Web page clustering into log file association mining and uses the cluster labels as Web page content indicators. Evaluations showed that the mined association rules of the log files are related to Web page contents. These rules may contribute to various concerns including Web user profiling, and Web construction improvement.