A Study on Implementation of Intelligent CAI

点击翻译按钮获取中文摘要

In the first year of the project, fundamental techniques of implementing Intelligent CAI systems were investigated by using the following methods, which we had already proposed before. They are "Multi-hierarchical Knowledge Model" used for representing a domain expertise, "Perturbation Method" used for diagnosing students' error origins and constructing student models, and a method to determine teaching strategy realizing various tutoring paradigms to give a hint, a counter example, an refutation, an explanation, a suitable problem, verification of accounts, etc.In the second year of the project, we proposed a method to generate "soft scenario", which plans content and expression of discourse between a student and the system pliably according to the student comprehension level, intention, request, etc. We also proposed a method to determine when and how to intervene the student learning process.Last year, we constructed a framework of intelligent CAI based on the above methods. Three systems, "ITS : Recognizing the context and guiding reduction processes for fractional calculation", "Thinking Tool : Improving Student's Reasoning Abilities" and "Intelligent CAI for English" have been implemented in the framework. These systems were exhibited publicly at the international conference "Advanced Research on Computers in Education".