Performance Enhancement of Web Server log for Distinct User Identification through different factors

Abstract: Nowadays millions of users daily interact with web sites around the world. Huge amount of data are being generated and these information could be very prized to the company in the field of understanding Customer’s behaviours. Web usage mining is relative independent, but not isolated category, which mainly describes the techniques that discover the user's usage pattern and try to predict the user's behaviours. Web usage mining is field of data mining which deals with the originality and study of usage patterns with use of Web log data. Specifically web logs in direction to advance web based applications. Web usage mining involves of three stages, Preprocessing, Pattern discovery and Pattern analysis. Work represent a Preprocessing methodology having Data Cleaning, User Identification which is key issue in preprocessing technique phase is to identify the unique web users. Traditional User Identification is founded on the site structure and user IPaddress by using some observed rules and based on the site topology which reduced the efficiency of result. In proposed work this problem is resolve through Distinct User Identification (DUI) Technique which is based on IP address, Agent, Referred pages, status on requested page and complete session time. It can be used in unique identification of users as well as through discovery of regular access process get better designing for site modification and business intelligence. Proposed work experiments get precious accurate result which improved the overall quality of preprocessing results which more beneficial for web usage mining techniques. Cause of using different factors get well-organized preprocessed data with more time consumption.

Keywords: Web usage mining, Preprocessing, User identification, Session time, Server log