Minerando Dados de Ambientes Virtuais de Aprendizagem para Predição de Desempenho de Estudantes

Abstract

Computer Engineering undergraduate students of Universidade Federal do Pampa face severe dificulties in introductory courses. In order to contributing on decreasing this problem, it started to be used a Course Management System to provide Learning Objects and formative evaluation resources. This system started to collect a huge amount of data about students’ behaviour and performance. In this scope, this paper describes the use of data mining algorithms in order to discover knowledge about data obtained in the Moodle Course Management System. Based on these discoveries, it will be possible providing useful information to professors and managers as a mean to creating preventive actions to aim increase students’ performance in summative evaluations.