Our proposed Hybrid Tutor provides a real-time assessment of student state that is richer than current approaches. It includes performance-based measures of actions and choices during a realistic simulation, as well as knowledge-based measures of student plans and explanations. It includes latency and self-assessment measures that provide information for an affect model of certainty and confidence. A Bayesian analysis takes these various measures and forms a student state model consisting of knowledge, skills, and affect. The student state model is continually updated, and influences the tutor's strategies so that it customizes interactions and instruction to the individual student. The ITS will be developed so that the main components (student model, domain knowledge, and tutor strategies) are reusable for other tutoring applications. The ITS will be plug-compatible with existing PC-based simulations through the use of software connectors. This will allow the tutor opportunities to gather input from the student and provide feedback and explanations. Evaluation of student actions in the simulation is performed by ontological reasoning. This reasoning is supported by a domain knowledge representation, a domain specific knowledge base backed by a standard upper ontology. Thus the tutor has an expert level active knowledge of domain concepts, rules and solutions. We expect that the proposed approach will provide improved training effectiveness and cost reduction. Improved training effectiveness is due to a richer student state model, more customized tutorial interactions, and more realistic training in simulations. This improved training can act as an adjunct to existing CTCs by providing advance training before rotation, and allowing refresher training after rotation when skills might otherwise deteriorate. Cost reductions will be due to reusable ITS components and knowledge for a variety of tutoring systems and plug-in connectors to existing and future simulations, thus leveraging millions of dollars of development cost.
Keywords: Student Model, Intelligent Tutoring System, Simulations, Bayesian Analysis, Ontology, Knowledge Base, Reasoning, Self-Assessment
