Authoring system for the development of ITS

Abstract

Suggests new paradigm for the development of intelligent authoring systems (IAS) for producing intelligent tutoring systems (ITS) with intelligent instruction. The ITS modules, independently generated by an IAS, consist of an expert module, a student module, a instructional strategy module, and an information interface module. The proposed authoring system separates out two knowledge networks with domain knowledge: a conceptual network and an instructional unit network, for the development of different types of ITS. These enable one to dynamically generate problems and plan a dynamic learning strategy. The paradigm accommodates an object-oriented approach in order to optimize the knowledge networks using inheritance among the knowledge units. The author can easily design the customizable student module according to his/her purpose, using the principle of inference to infer the student's knowledge status and diagnose the student's responses using the linguistic variable of fuzzy theory with the concepts of concentration and dilation. Dynamic instructional planning separates the knowledge network at the abstraction level, and proposes an algorithm that produces a curriculum plan. >