Preprocessing the web server logs

Abstract

Data preprocessing is an important activity for discovering behavioral patterns. The analysis of web logs is an essential task for System Administrators to safeguard adequate bandwidth and to maintain server capacity on their business websites. A web Log file represents user activities occurring over a period of time. Web log files offer valuable insight into the effective usage of the web site. It helps maintain an account of the actual usage in a regular working system as compared to the virtual setting of a usability lab. This research paper focuses on the preprocessing techniques implemented on a specially designed Web Sift (WebIS) tool on an IIS web server and also proposes some efficient heuristics and techniques