A Case-based Reasoning Approach to Intelligent Tutoring Systems

In Phase I, we will investigate a case-based reasoning (CBR) approach to intelligent tutoring systems (ITSs) and explore the feasibility of semi-automated/automated course authoring techniques to speed collection and structuring of domain knowledge. We will develop a prototype example-based ITS, enhanced with multimedia, within a specific Navy instructional domain and a generic ITS architecture applicable in many other domains. The use of an ITS for individual or team training will achieve many of the same benefits as one-on-one instruction. Additionally, an ITS can promote instructor productivity, cope with the higher skill requirements for trainees, provide tailored instruction and remediation, while allowing flexibility in teaching methods. When ITSs are enhanced with interactive multimedia, students demonstrate greater motivation during training and greater retention of course material. Thus, an ITS provides instruction that is at once cost-effective and potentially superior to traditional classroom teaching methods. An example-based approach offers the further benefits of automatically or semi-automatically generating the student's mental model and of largely circumventing the difficult and time- consuming process of constructing an explicit expert mental model. Furthermore, the more examples presented to the student, the more likely it is that the student will be able to apply concepts in new situations. We have used this CBR approach with much success in other training domains.