Abstract
Adaptive web interfaces help users to perform tasks in an application and also construct a user's preferences model to best serve them in the future. This paper presents usability evaluation of adaptive web interface which focuses on how users can learn to achieve their goals. First, we present our adaptive Web interface using a Bayesian networks approach. Then, a formal GOMS model approach was applied to the evaluation of our user interface for a specialized web application. The evaluation shows that the adaptive user interface was more comfortable than the fixed user interface.