An approach for dynamic student modelling of learning styles

Knowing students' learning styles can contribute highly in supporting students in learning by making them aware of their learning styles and/or presenting them with personalised learning material and activities. In this paper, we propose a dynamic student modelling approach, which monitors students' behaviour in a course and notices once the learning styles stored in the student model do not reflect students' behaviour in the course any more. The proposed approach aims, on one hand, at updating the information in the student model as soon as possible once differences in students' behaviour were detected but, on the other hand, also considers deviations in students' behaviour. In an experiment with data from 75 students, appropriate parameter settings and the suitability of all considered components were tested, showing that the proposed dynamic student modelling approach yields good results. By using this dynamic student modelling approach, a system is therefore able to revise the information in the student model when required and thus enable systems to provide students with support, learning material and activities they currently need.