Data Pre-Processing on Web Server Access Logs of University for User Interaction Patterns

Abstract

In the current digital Era, websites are developed and organized into multifaceted in nature. It is essential to distinguish user sessions/intent and browsing behavior from logs in order to recommend appropriate content for the web designers and administrators. This paper focuses on data preprocessing of the weblogs received from Kannada University Hampi, Vidyaranya Karnataka state are cleaned viably by applying various pre-processing methodologies. The work identifies the superior quality of data to discover user interactions, user sessions, the specific web pages, and the regularly visited Uri’s, most visited pages, most time spent on pages and incorrect webpages served to users. These pre-processed webserver access log files will be utilized to discover patterns, fine grained analysis and study. This paper also focuses on challenges of log file analysis.