Exploiting Knowledge Representation in an Intelligent Tutoring System for English Lexical Errors

Generally, Intelligent Tutoring Systems (ITSs) rely heavily on the hand-construction and design of the domain knowledge provided for learners. In this paper we attempt to illustrate a central role that knowledge representation can play in automating crucial aspects of ITS design and implementation. We focus on the treatment of English learners’ lexical errors. Specifically, we describe a system which exploits highly constrained dialogue with the learners in order to allow a simple algorithm to retrieve the relevant lexical knowledge that the learner needs to remediate the error at hand. For this, we propose a Diagnosis, Interaction and Treatment (DIT) Model for ITS. The entire system relies upon a knowledge representation system (InfoMap) whose structured encoding of English lexical information makes it possible to (1) initiate the relevant dialogue with learners when the system is not sure about learners’ intentions, (2) trigger the appropriate lexical knowledge based on their responses, and (3) automatically generate practice and test exercises on this knowledge. Crucially, inferences from the dialogue and automatic practice exercise generation rely on a very general algorithm and the structure of the English knowledge representation, regardless of the ability of the learner. Thus, the system automates what must traditionally be created by hand through human experts.