Intelligence on the Web?

This paper describes a courseware system delivered over the World Wide Web for transforming existing video-taped courses. The on-line courses take advantage of the interactivity and the individualized instruction of intelligent tutoring systems. They use synchronized audio and HTML slides to present material to the student. An overlay student model records a student’s ability on each topic in the system, determined by which slides the student has seen and quizzes he has taken. This model is then used to help guide the student through the material, to provide interactive and adaptive quizzes, and to prefetch course material before it is explicitly requested, reducing the delays seen by the student. The first goal is accomplished by using adaptive hypertext techniques including adaptive navigation support and adaptive presentation. The second goal is achieved by dynamically constructing quizzes at the correct level of difficulty for the student. The third goal is accomplished by predicting the next actions the student will take, and preparing for them ahead of time. In this paper we discuss initial efforts to implement a student model for the kinds of web-based courses supported in the system. We also discuss how the student model is used to accomplish the three goals. We conclude by providing a brief description of the first student encounters with the system.