Análise da evasão estudantil com estudo de caso o Ifes campus Cachoeiro de Itapemirim utilizando recursos do Machine Learning

Abstract

ABSTRACT: Student dropout is a significant challenge at different levels of education, and its direct impact affects both students and institutions across the board. This is due to the considerable investment made in infrastructure and teaching staff, which ends up resulting in losses when dropout occurs. Given this reality, the objective of this work is to carry out a detailed analysis through a case study on the dropout of students in undergraduate courses offered by the Federal Institute of Espírito Santo, at Campus Cachoeiro de Itapemirim. In which they were prepared and normalized to be divided into training and test bases to then be applied and identify among the Machine Learning algorithms DecisionTreeClassifier, RandomForestClassifier and MLPClassifier which has the best performance to be evaluated. The RandomForestClassifier Algorithm stood out between and the three algorithms and was applied to later metrics such as accuracy and precision to understand the assertiveness and performance of the algorithm in relation to the base, thus determining the potential of students to remain.