View article

As a generalized framework for intelligent tutoring systems (ITSs), the Generalized Intelligent Framework for Tutoring (GIFT) integrates various types of ITS modules. Among all the modules, conversational ITS modules might be the most complex one, since it adds complex conversations to any type of ITS environment. Each type of module in ITS may have its specific considerations on domain modeling. This chapter presents our experience in domain modeling on “AutoTutor”, which has been integrated as a conversation module in GIFT.

The domain model has long been considered as one of the major components of an ITS. For example, Elsom-Cook (1997) assumed that the structure of an ITS is around a “trinity” of components:“the knowledge about the domain being taught, about how to interact with the learner, and about the student.” Murray (1999) described four models in authoring ITSs: the interface model, the domain model, the teaching model, and the student model. In the preface of the first book of this series, Sottilare, Graesser, Hu and Holden (2013) indicated that domain models in an ITS contain “the set of skills, knowledge, and strategies of the topic being tutored.” They added that a domain model “normally contains the ideal expert knowledge and also the bugs, mal-rules, and misconceptions that students periodically exhibit.”