Consistent Evolution of Student Models by Automatic Detection of Learning Styles

One of the most important features of adaptative e-learning systems is the personalisation according to specific requirements of each individual student. In considering learning and how to improve student learning, these systems must know the way in which an individual learns. In this context, we introduce a new approach for consistent evolution of student models by automatic detection of student learning styles. Most of the work in this field presents complex and inefficient approachs. Our approach is based on learning styles combination and dynamic correction of inconsistencies in the student model, taking into account the non-deterministic aspect of the learning process. Promising results were obtained from tests, and some of them are discussed in this paper.