Abstrato

An Exclusive Survey on Web Usage Mining For User Identification

Satpal Singh, Vivek Badhe

Web mining has been explored to a vast degree with different techniques that has been proposed for a variety of applications. Most research on web mining has been done on “data-centric” point of view and a few works has been done on “user-centric” view. In this paper we explored the definition of web-user and also examined the various dimensions of temporal web mining. In web usage mining, web log file plays an important roll, some fields of the web log file are used frequently and some other fields like DateTime are rarely used. They are in advertently removed in the cleaning session. This paper emphasizes the importance of these fields, which can be very useful for user identification process. We study in particular the behaviour of web-usage data over a period of time. In this paper we have suggested the different view point for finding the web-user on the basic of temporal approach. Such kind of analysis could be useful for target marketing based on time or for web services optimization.