An XML-based architecture for adaptive Web hypermedia systems using a probabilistic user model

Web based hypermedia systems are becoming increasingly popular as tools for user driven access to information and services. The paper presents an architecture for the development of Web based adaptive hypermedia systems. The architecture uses weighted graphs of XML documents to describe the application domain and a probabilistic model to adapt the Web site content generation and presentation to the user's behaviour. The user's behaviour is modelled using a probabilistic model and the most promising profile, that is a “view” over the application domain, is dynamically assigned to the user, using a discrete probability density function