Abstract—Web usage mining is research area in web mining. Web mining is an activity that focuses to discover new, relevant and reliable information and knowledge by examining the structure, content and usage of web. The major focus is on learning about web users and their interaction with websites. Web log files generated on web servers are used in order to extract web usage of different users. There are three types of web repositories: web server log, proxy server log, browser log. Analysing web logs for usage can not only provide important information to websites developers but also help in creating adaptive web sites. In this paper we discuss various sources of information for WUM, Methodology of web usage mining techniques which involves Data collection , Data pre-processing , knowledge discovery and knowledge analysis. Various applications of WUM are personalization, prefetching and caching, support to design and E-commerce. Major application of web usage mining is to predict future accesses. Thus, the result obtained after web usage mining can be used to improve the performance of prefetching and caching.

Index terms—web usage mining, methodology, preprocessing, clustering, classification, applications

Cite: Mehak Jain, Mukesh Kumar, and Naveen Aggarwal, "Web Usage Mining: An Analysis," Journal of Emerging Technologies in Web Intelligence, Vol. 5, No. 3, pp. 240-246, August 2013. doi:10.4304/jetwi.5.3.240-246