Simultaneous assessment of teams in collaborative virtual environments using Fuzzy Naive Bayes

In the recent years, computational architectures have been proposed to allow teams assessment in collaborative training based on virtual reality. In the virtual environments, procedures are performed by a team of professionals acting simultaneously, as in real surgical rooms. It is important to verify if the group performed the procedure correctly or not. The assessment systems utilizes user and users data from the execution of the virtual procedure, generated by the virtual reality system, to be compared with predefined classes of performance. Previous approaches basically used fuzzy rule based expert systems and presented some problems with respect to calibration which was performed in phases. In this paper, we propose a new approach based on Fuzzy Naive Bayes to perform the calibration in a single phase, without lost of accuracy in the assessment of the performance.