A Constraint-Based Expert Modeling Approach for Ill-Defined Tutoring Domains.
Angela Woods, Brian Stensrud, Robert E. Wray, Joshua Haley, Randolph M. Jones · The Florida AI Research Society · 2015
We introduce an approach for representing and diagramming machine-readable expert models for intelligent tutoring systems (ITSs) and virtual practice/training environments. Building on previous work on constraint-based expert models, we address the challenges of implementing ITS features within complex, ill-defined learning domains. Our constraint-based expert model (CBEM) can be used, in conjunction with a real-time interpreter (the Monitor), to make accurate, real-time observations of student behavior in illdefined domains. These observations can be used for in-situ feedback and dynamic, individualized experience tailoring and instruction. In this paper, we detail the structure and elements of the CBEM, describe instances of successful application to several training domains, and introduce current and future research to extend and improve the paradigm.