The representation of domain knowledge in intelligent training systems

The ESPRIT project 2615 ITSIE, aims to utilise ITS techniques to develop a generic framework for the development of training systems. A major aspect of this project is the use of the qualitative modelling techniques to meet the needs of current training systems. The project has recognised the importance of clearly establishing the types of skills and `expertise´ required of the trainee. The authors believe that this expertise exists in various forms and should be supported by appropriate representational tools. A systematic classification of domain knowledge for intelligent training is presented as used within the ITSIE project. Based on earlier studies on the classification of the behaviour of humans in industrial environments, it is proposed that the domain knowledge be classified into two main categories: domain expertise and domain simulation. The proposed classification and tools provide a coherent way of representing domain knowledge in ITS and allow the use of multiple models in a model switching framework for effective training