Web Usage mining for Predicting User Access

Abstract — In current trend of computer science, there are lots of software available for data mining, predictive analytics, and business analytics. In every sort of software where expectation comes for prediction then there is need for a web log files of user visit to particular website stored at server side. The interesting information for extracting knowledge from such huge data demands for new logic and new methods. In this paper work, we proposed a web usage mining method with a well-known mining tool software. At first, we are having a log file from KDD repository then we performed deletion on the redundant information from file and then we mining log file which has been submitted with mining tool, then going to get the customized form of access done by different user to visit the website by performing processing and analyzing phase on log file, and mine for so called unusual rules, and suggesting rules the as reference for the decision making and construction of website. At the end, we collected experimental result and analysis on result show that, applying web usage mining, will can look of frequent model which user visits the website, manage to optimize the website structure and recommends for users