Finding the most evident co-clusters on web log dataset using frequent super-sequence mining

Abstract

It is important to mine the weblog dataset to find interesting and helpful information. There are three kinds of mining on weblog data which are web usage mining, web structure mining and web content mining. In our research, we are going to investigate web pages structure and find the most evident groups of users and web pages. Nowadays, big data is everywhere. Facing huge amount of web logs, it is not always necessary to group all the users in a web log dataset into different clusters, sometimes, finding out the major dominant user groups and the corresponding web pages is more important. In this paper, we are going to investigate a new way to search the most evident co-clusters of users and the corresponding web pages in the web log dataset using frequent super-sequence mining technique. Through experiments we find interesting results.