PREDICTING USERS IN BROWSING BEHAVIOUR FOR WEB USAGE MINING USING FCM CLUSTERING

World Wide Web is a huge storehouse of web pages and links. It offers large quantity of data for the Internet users. The growth of web is incredible as around one million pages are added per day. Users' accesses are recorded in web logs. Web usage mining is a kind of mining techniques in logs. Because of the remarkable usage, the log files are growing at a faster rate and the size is becoming very large. This leads to the difficulty for mining the usage log according to the needs. This provides a vast field for the researchers to provide their suggestion to develop a better mining technique. Then the researchers propose the hierarchical agglomerative clustering to cluster users' browsing patterns. The provided prediction by two levels of prediction model framework work healthy in general cases. On the other hand, two levels of prediction model experience from the heterogeneity user's behavior. In this paper, the author enhances the two levels of Prediction Model to achieve higher hit ratio. This paper uses Fuzzy Possibilistic algorithm for clustering. The experimental result shows that the proposed techniques results in better hit ratio than the existing techniques.