Open student model for blended training in the electrical tests domain

Abstract

Electrical tests are important because they anticipate problems. When they are erroneously performed, human accidents or equipment damage can occur; thus, efficient training is mandatory. Traditional training under supervision of human instructors has proved to be successful; but it is costly and takes a long time. We need to complement traditional training with computers systems which adapt the training to particular trainee and shorten the training time. We have defined a blended learning model to support adaptive and distance training for complementing traditional training. We have defined an open trainee model to represent how much trainees know and how they feel. The model is used by instructors to adapt instruction, and by trainees to know what needs to be reinforced. The trainee model is represented by Bayesian networks and the instruction is presented via a virtual reality system. This paper presents the open trainee model and current results on using it.