A probabilistic adaptive hypermedia system

Adaptive hypermedia systems allow user-driven access to information and services and content personalization. This paper presents the design and development of a probabilistic adaptive hypermedia system that uses a probabilistic approach for user modelling and presentation adaptation. The application domain is modelled along three different adaptivity dimensions: the user´s behaviour (browsing activity), technology (network and user´s terminal constraints), and external environment (time, location, language, socio-political issues, etc.). The user´s behaviour is modelled using a probabilistic approach, and the most promising profile, i.e. a “view” over the application domain, is dynamically assigned to the user, using a discrete probability density function. A modular architecture, implementing both the authoring phase and the run-time support of the adaptive hypermedia system, is also presented