Auxiliando o desempenho de alunos com tendência a evasão na educação a distância utilizando técnicas de mineração de dados e sistemas multiagentes

Abstract

The distance education is a form of education that has as main tool the Virtual Learning Environment (VLE), where teachers, students and tutors can interact directly or indirectly. A common problem in EAD is the large amount of students who end up quitting school. However it is necessary to follow those students periodically to change this scenario and leading them to success in their studies. With the great increase in the amount of information generated by students interacting in VLE’s, data mining comes as an opportunity to discover patterns on the behaviors of students, and with to be able to identify early students who tend to have a good or bad academic performance. This paper describes the development of a Multi-agent system module (MAS) developed by the Study Group in Software Engineering and Multi-agent Systems (GESMA) with the purpose to monitor the behavior of the students and identify as soon as possible when they tend to escape, engaging the other agents in the system to aid the student in their school success.