Evaluation Models for Expert Systems in Military Training

This paper focuses on evaluation models for the dominant application of expert systems in training, that is, Intelligent Tutoring Systems (ITS) or Intelligent Computer-Assisted Instruction (ICAI). Here we provide a new evaluation model for such applications. Since intelligent CAI shares some common attributes with traditional CAI, evaluation technology approaches to the traditional will be integrated into our model. A case study of the model designed for an ICAI program in weather is provided. A wide range of evaluation techniques would be included, for instance, both quantitative and qualitative data collection and analyses. Instrumentation and data collection strategies are suggested along with student and instructor questionnaires. Finally, we make suggestions for measuring intelligence in ICAI.