Web usage mining on the sequences of clicking patterns in a grid computing environment

Abstract

Analyzing user's browsing sequences is an important task of web usage mining. It can help the web administrators or designers improve the web structure or tune the performance of the web servers. Mining on the sequences of such clicking patterns (MSCP) can be considered as a data mining task. MSCP is usually a costly task due to its considerable amount of time for computation and storage for archiving a huge amount of data. Running MSCP becomes inefficient or even not workable on a computer with limited resources. This paper explores the deployment of MSCP in a distributed grid computing environment and demonstrates its effectiveness by empirical cases. The use of a grid-environment and associated experimental results are presented and discussed.