Discovery and Analysis of Usage Patterns for Web Personalization

Abstract

In this paper, we present a community discovery method based on information extraction from user sessions in order to find usage patterns. We have characterized the overall access from each page in the web site and have detected the pertinent users within communities using the potentially useful information available in different user sessions. Then we compare the proposed method with the random walk method which is one of the most popular graph clustering methods that gives a good partition of weighted and overlapped graphs. As a result, the proposed method helps us understand the behavior of the user in the web and improve access modes to information.
DOI: 10.17762/ijritcc2321-8169.150232