Agent model for problem solving towards simulation-based learning environment

摘要

In this paper we propose a methodology for the design of a tutor agent based on constraint model The function of cur tutor agent is to interpret the student's constraint violations and draw a behavioral schema of the student's knowledge transfer skills on neural networks concepts. This enables us to generate from a set of constraints desirable functions such as, task generation, hints, and domain application metaphors related to the student's learning goals. We present the fundamental issue of how to represent a student knowledge transfer skills in reactive tutor agent.