Abstract

Web usage mining is obtaining the interesting and constructive knowledge and implicit information from activities related to the WWW. Web servers trace and gather information about user interactions every time the user requests for particular resources. Evaluating the Web access logs would assist in predicting the user behavior and also assists in formulating the web structure. Based on the applications point of view, information extracted from the Web usage patterns possibly directly applied to competently manage activities related to e-business, e-services, e-education, on-line communities and so on. On the other hand, since the size and density of the data grows rapidly, the information provided by existing Web log file analysis tools may possibly provide insufficient information and hence more intelligent mining techniques are needed. There are several approaches previously available for web usage mining. The approaches available in the literature have their own merits and demerits. This paper focuses on the study and analysis of various existing web usage mining techniques.