Continuous Evaluation in Training Systems Based on Virtual Reality Using Fuzzy Rule Based Expert Systems

1 Liliane dos Santos Machado, Department of Informatics, Universidade Federal da Paraíba, Cidade Universitária s/n CEP 58.051-900 João Pessoa – PB Brazil, [email protected] 2 Ronei Marcos de Moraes, Department of Statistics, Universidade Federal da Paraíba, Cidade Universitária s/n CEP 58.051-900 João Pessoa – PB Brazil, [email protected] Abstract  The approach of continuous evaluation is an important tool in the learning process. However, only recently it was applied in training based on virtual reality. This paper presents a methodology of evaluation that uses the continuous evaluation approach to create an user profile from his several training. This information can be used to improve the user performance in the execution of the task. The methodology proposed is given by the union of classical statistics tools, fuzzy measures and a fuzzy rule based expert system (FRBES) to construct an individual profile for each trainee. Statistical tools, fuzzy measures and specific data of training provide information to the FRBES for the profile composition throughout training process. This new approach is a diagnostic tool that enables a trainee to understand the areas in which he presents difficulties and to concentrate to solve them.