Knowledge Acquisition for Intelligent Tutoring Systems

Intelligent Tutoring Systems require three different kinds of knowledge: the domain to be taught, techniques of teaching, and characteristics of the student. As such it is worth considering the extent to which knowledge acquisition techniques for conventional expert systems can be applied, and in particular those which aim to semi-automate the process. This paper considers some of the special features of knowledge for ITS that make a direct use of the knowledge acquisition techniques inappropriate. A particular kind of knowledge representation technique, the applicable rule is presented and its implications for knowledge acquisition noted. By considering the results of using a knowledge acquisition interface developed for obtaining information about second language learning and using the concept of applicable rules, the issues that subsequent knowledge acquisition systems for ITS should address are presented.