EVALUATION OF TRAINING EXECUTED BY WEB USING MULTILAYER PERCEPTRON BASED SYSTEMS
Santos Machado, Ronei Marcos de Moraes · 2006
Abstract ⎯ This paper presents a new methodology of evaluation of training executed by Web. To evaluate the user’s performance it is necessary collect data from training. Dedicated plug-ins are used to collect information about the different variables of user’s training. Some automatic evaluators use expert systems to perform evaluation using as input statistical models and statistical tests. However, in some applications, it is difficult to obtain knowledge from an expert and the data collected from user’s interaction cannot be adequate to classical statistical distributions. To solve these problems we propose an intelligent evaluation procedure that allows classifying a trainee learning using a Multi-Layer Perceptron neural network based system. The user can be classified into classes of learning giving him a real position about his performance, through the reports of performance. That reports can help user to improve your performance in execution of real procedure.