A Qualitative Comparison of Techniques for Student Modeling in Intelligent Tutoring Systems

Abstract

Intelligent tutoring systems (ITS) are interactive learning environments based on instruction assisted by computers. The intelligence of these systems is largely attributed to their ability to adapt to a specific student during the teaching process. In general, the adaptation process can be described by three phases: (i) getting the information about the student, (ii) processing the information to initialize and update a student model, and (iii) using the student model to provide the adaptation. In this paper we studied aspects related with student modeling (SM) in intelligent tutoring systems. First we make a qualitative comparison of two techniques: Bayesian networks (BN) and case-based reasoning (CBR) for SM. We apply both techniques to a case study concerning the development of an ITS for e-learning in the medical domain. Finally, we discuss the results obtained