Discovering and visualizing temporal-based Web access behavior

Discovering and understanding Web users' surfing behavior are essential for the development of successful Web monitoring and recommendation systems. In this paper, we propose a Web usage mining approach for the automatic discovery and visualization of temporal-based Web access behavior of individual users by mining client-side logs. The proposed approach is based on a Web usage lattice model which represents a hierarchy of Web access activities. To describe such Web access activities, we incorporate fuzzy logic to represent real life temporal concepts such as morning, afternoon and evening, and meaningful Web categories such as news, sports and chat. Based on the lattice, temporal and association behavior patterns can be extracted and visualized.