Web Log Pre-Processing For Web Usage Mining

In our proposed system logs will be pre-processed and generated in graphical or tabular format according to the requirement of client (e.g.-manufactures) Therefore even though the system is complex but user session sequences are generated with less time and greater precision. Web Usage Mining is to discover and extract useful information. It helps in better understanding and better serve the needs of web-based applications. The data should be preprocessed to improve the efficiency and ease of the mining process. The proper analysis of web log file is beneficial to manage the websites effectively for administrative and user’s perspective. Many a time the backward referencing used to track the reachability of the pages which consumes time and generates complete path from the root node even if the link has come from some other server page. This problem can be using two-way hash table structure as Access History List. In our system, session identification is done using AHL by considering immediate link analysis, backward referencing without searching the whole tree representing the server pages. Based on this study, it can be concluded that the system is complex but user session sequences are generated with less time and greater precision.