User modeling for intelligent interfaces in e-learning.

The emergence of the accessible knowledge society for all underlines the need for all to acquire the necessary knowledge and skills for inclusion. One way to do so is through e-learning, which itself should be accessible to all intended users. The mere provision of knowledge systems is not enough, since the need is for active and accessible learning that engages the participants effectively. This study explores the hypotheses that the solution is to be found in systems that: (i) support rather than replace the tutor, (ii) are accessible to the intended populations of users, (iii) can be adapted to the needs and individual characteristics of diverse users, (iv) are smart enough to adapt in real time to changing user needs, (v) reflect best practice in learning psychology, (vi) provide a high quality student experience, (vii) provide a high quality tutor experience and (viii) use valid student modeling. These hypotheses are evaluated through the five case studies. Accessibility and student modeling turnout to be the weakest points. All the other features are well represented in the case studies. None placed any kind of substantial emphasis upon accessibility. Only one of the case studies (case study 2; Cognitive Tutor Authoring Tools) makes a substantial effort in user modeling, being based upon the generic, cognitive model ACT. Even then, the focus tends to be on the typical or average user and does not address the problems of user diversity. Clearly, accessibility and user modeling need much more work in e-learning.

4th international conference on 'Universal access in human-computer interaction' UAHCI 2007 held as part of HCI International 2007.
Beijing, China, July 22-27, 2007
Proceedings, Part III