A New Approach for Modeling and Discovering Learning Styles by using Hidden Markov Model

Abstract

Adaptive learning systems are developed rapidly in recent years and the "heart" of such systems is user model. User model is the representation of information about an individual that is essential for an adaptive system to provide the adaptation effect, i.e., to behave differently for different users. There are some main features in user model such as: - knowledge, goals, learning styles, interests, background... but knowledge, learning styles and goals are features attracting-researchers' attention in adaptive e-learning domain. Learning-styles were surveyed in psychological theories but it is slightly difficult to model them in the domain of computer science-because learning styles are too unobvious to represent them and there is no solid inference mechanism for discovering-users' learning styles now. Moreover, researchers in domain of computer science will get confused by so many psychological theories about learning style when choosing which theory is-appropriate to adaptive system. - In this paper we give the overview of learning styles-for answering the question "what are learning styles?" and then propose the new approach to model and discover students' learning styles by using Hidden Markov model (HMM). HMM is such a powerful statistical tool that it allows us to predict users' learning styles from observed evidences about them.