Abstract-Managing a dialogue between a student and an intelligent tutoring system is a challenging problem for many applications. It has often been argued and demonstrated that adaptive dialogues between a user and a computer can be generated automatically, using automated planning techniques to plan speech acts. To date such plan-based dialogue generation approaches have relied on deterministic planning algorithms. Consequently they can only handle sequential dialogue structures. In this paper we describe a new approach for automatically planning more general tree-like dialogue structures, by using a nondeterministic planner with incomplete knowledge and sensing. Our approach takes into account incomplete information about the user's knowledge by including queries that the computer can ask to the user to gather missing information that is necessary for an effective feedback. We illustrate our system with an application to an intelligent tutoring system for medical diagnosis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.