Abstract: Enormous amount of informationís are gathered and viewed through world wide web by different users. The user practices their views by entering hypertext credentials by internet with a large repository of web pages and web usage mining process is essential for efficient web site management, personalization, business and support services and network traffic flow analysis, etc., web page contains images, text, videos and other multimedia and web log file holds the information of the user accesses in the websites. The log file shall have some noisy and ambiguous data which may affect the data mining process and large quantity of web traffic should be handled effectively to acquire desired information. So the log file should be preprocessed to improve the quality of data. Preprocessing consists of data cleaning and data filtering, user identification and session identification. Two sets of log files are collected and processed to obtain experimental results. This study presents a framework for user and session preprocessing and clustering with Hidden Damage Data algorithm (HDD) and also analyzes the navigational behavior of users through an enhanced Conviction Frequent Pattern Mining Algorithm (CFPMA) to identify frequent patterns in web log data. The experimental result shows that the proposed technique achieves low execution time and higher accuracy when compared with the other existing methods.
P. Senthil Pandian, K. Karthikeyan and K.N. Sivabalan, 2016. Advancement in Analysis of Preprocessing and Frequent Patterns in Web Usage Mining. Asian Journal of Information Technology, 15: 3407-3413.