ANÁLISE DA EVASÃO DE ALUNOS NO CENTRO UNIVERSITÁRIO PARTICULAR UTILIZANDO MODELOS DE MACHINE LEARNING

Abstract

Student dropout in higher education is a critical issue, affecting the structure of institutions and the academic lives of students. This study investigates the applicability of Machine Learning models in predicting student dropout at Centro Universitário Particular. Using data, the study proposes the identification of patterns and characteristics that influence the evasion decision, applying advanced predictive analysis techniques. The methodology employs collection, exploratory analysis, data pre-processing, and the development and evaluation of predictive models. The results indicate significant results that influence dropout rates, with the implementation of the predictive model functioning as a decision support tool for student retention policies.