Martin, J. & VanLehn, K. (1993). OLAE: Progress toward a multi-activity, Bayesian student modeller. In S. P. Brna, S. Ohlsson, & H. Pain (Eds.), Artificial Intelligence in Education, 1993: Proceedings of AI-ED 93 (pp. 410-417). Charlottesville, VA: Association for the Advancement of Computing in Education.

There are many obstacles to effective student modeling, in this article, the authors address three. A student modeling system should: (a) analyze data in a statistically sound, defensible manner, (b) augment data on a person's performance while they work with data from other tasks, and (c) provide assessments at multiple grain sizes. The authors present OLAE, a computer assistant to a human assessor that collects data about problem solving in elementary physics, analyzes that data with sound, probabilistic methods, and flexibly presents the results of analysis.

For a pdf of the paper, click here (216KB).

This paper won the "best paper" prize of the conference.