Modelling the Student to Individualise Tutoring in a Web-Based ICALL

In this paper we describe the techniques used for student modelling in an adaptive Webbased Intelligent Computer Assisted Language Learning (ICALL) system. The system teaches the domain of the passive voice of the English language and is called Web Passive Voice Tutor (WebPVT). It dynamically models a student's progress in learning and is able to provide individualised tutoring and advice tailored to a student's knowledge level and individual weaknesses. The initialisation of the student model is based on a multi-dimensional stereotype approach. However, after the individual student has interacted with the system sufficiently, the initial values provided by the stereotype are overwritten to reflect the individual student. This allows the system to adapt to each particular student.