EvolveDTree: Analyzing Student Dropout in Universities

Abstract

Brazilian society suffers constant financial losses when higher education students disassociate from universities without completing the degree program in which they were enrolled. This is especially true when the institutions are funded through public resources. In order to minimize evasion losses, socioeconomic policies and programs were created to assist and support actions seeking to maximize the number of students that graduate in a suitable program time. This work presents a methodology that aims to predict evasion by using machine learning. Our approach was able to classify student abandonment with an average f-score and accuracy results above 95%. Our approach combines a decision tree alongside a genetic algorithm and cluster stratified sampling. The results obtained show that students with a Grade Point Average (GPA) below 5.79 and that have been enrolled for more than a year require careful monitoring because they tend to exceed the program duration time or abandon it. Furthermore, approximately one-third of all identified dropouts students occurred in the first year.