Abstract

Web Usage Mining deals with the understanding of user behavior while interacting with the website by using various log files The whole process of Web Usage Mining gets completed in three phases namely Data Preprocessing Pattern Discovery and Pattern Analysis Data Preprocessing is important because it takes 80 of the time of the whole process of Web Usage Mining Data Preprocessing involves Data Cleaning User Identification and Session Identification In Session Identification we find out the set of pages visited by a user within the duration of one particular visit to a website also called as Sessionization In paper 1 we proposed a new method for session construction As the size of log files are very large so there is a requirement of an approach for Session Identification by which processing time of our proposed method will be reduced to a great extent In this paper we used Map-reduce method to calculate sessions in which we combine both time and user navigation method This approach is faster than the existing approach because we have performed the whole process in distributed environment