ABSTRACT

As the information available on World Wide Web is growing the usage of the web sites is also growing. Since each access to the web pages are recorded in the web logs it is becoming a huge data repository which when mined properly can provide useful information for decision making. The designer of the web site, analyst and management executives are interested in extracting this hidden information from web logs for decision making. In this research paper we proposed a method to categorize the users into faithful, Partially Impatient and Completely Impatient user, page wise so that study of user behavior can be easier. To categorize the user we proposed one new information in the web log that represent each instance of refreshing. We used the markov chain model in which we treated the clicking of Refresh button as another state i.e. Refresh State. We derive some theorem to study each type of user behavior and show that how do users behavior differ.

Keywords: Adaptive web sites, Markov chain model, Pattern discovery, Transition probability, Web mining.