Time- Driven Adaptive Web Personalization System for Dynamic Users

Web data has been increased to many folds in the recent past. Study of web data for collecting relevant information is indispensable for providing efficient web services. However, web designers are facing difficulty in organizing their site to meet the user's demands. Web personalization solves this problem to a certain extent. This paper presents an adaptive web personalization model which classifies the user's browsing behavior and predicts their interested areas. The proposed model is designed with a two tier architecture in which the first tier clusters the user's navigation patterns, according to the timing attributes extracted from the log file. In the second tier, the user's interested areas are predicted using a classifier. The proposed model is validated through experiments.