An approach to webpage prediction method using variable order markov model in recommendation systems

With the continuous increase of web applications and services, web usage data have been overloaded and handling that dynamic web information is very challenging task. Personalized web recommender systems are evolving to provide better and tailored experiences for online users than ever before. The personalization technique is carried out for the each individual user by considering their interests and search behavior stored in the web server access logs. Recently a variety of recommendation systems to predict user future request has been proposed, but the quality of these system results only low prediction accuracy. Hence this paper presents a new framework for effective recommendation system to reduce the searching time of the user and to reach the user’s future intention (request) webpage with the improved prediction accuracy by integrating fuzzy c-means clustering and variable order markov model recommendation system. Experimental setups are carried out initially by applying preprocessing technique on the web log followed by fuzzy c-means clustering process to identify the similarity patterns. Finally web page recommendation is performed using variable order markov model to predict the user’s next web page access by reducing the search time and better prediction accuracy.