System Architecture

Figure 1 depicts a general architecture for Web personalization based on usage and content mining. The overall process is divided into two components: the offline component which is comprised of the data preparation and specific Web mining tasks, and the online component which is a real-time recommendation engine. The data preparation tasks result in aggregate structures containing the preprocessed usage and content data to be used in the mining stage. Usage mining tasks can involve the discovery of association rules, sequential patterns, pageview clusters, or transaction clusters, while content mining tasks may involve feature clustering (based on occurrence patterns of features in pageviews), pageview clustering based on content or meta-data attributes, or the discovery of (content-based) association rules among features or pageviews. In this paper, we focus on the derivation of usage profiles from transaction clusters, and the derivation of content profiles from feature clusters. In the online component of the system, the recommendation engine considers the active server session in conjunction with the discovered patterns and profiles to provide personalized content. The personalized content can take the form of recommended links or products, targeted advertisements, or text and graphics tailored to the user's perceived preferences as determined by the matching usage and content profiles.

Figure 1: A general framework for automatic personalization based on Web Mining

2000-08-14