Abstract

We present in this paper the architecture of MetaTutor, an intelligent tutoring system that teaches students meta-cognitive strategies while learning about complex science topics. A more in-depth presentation of the micro-dialogue component in MetaTutor is provided. This component handles the meta-cognitive strategy of subgoal generation. This strategy involves subgoal assessment and feedback generation. We present a taxonomy-driven method for subgoal assessment and feedback. The method yields very good to excellent human-computer agreement scores for subgoal assessment (average kappa=0.77).

Description

Proceedings of the Twenty-Third International Florida Artificial Intelligence Research Society Conference (FLAIRS 2010)

Links and resources

Tags