Identificação de Perfis e Padrões de Participação dos Estudantes de Cursos a Distância na UFRN por meio de Mineração de Dados

Abstract

The Interest in data mining theme emerged from the concern of having information about the student's performance during the course in online learning. Moodle has several types of reports, however it provides no data about the students activities and profile of students performance for teachers decision-making. The research aimed to apply a data mining methodology to identify the profiles and participation patterns of students in distance education courses, resulting in predicting the chances of approval for each student. It was developed an applied project based on an adaptation of KDD for educational data mining in Moodle platform. The analyzed data was extracted from a subject of online courses from the SEDIS database with 497 enrolled students.The result was the formation of four clusters designating the profiles and variety of student participation, namely: active,median, inconsistent and absent students. This study is a contribution of computer area for education because it allows the teacher to have access to student performance profile during the subject and then take some preventive measures to avoid disapproval or locking of subject by the students.