Desenvolvimento de um modelo de Inteligência Artificial com Scikit-Learn para Predição de Reprovação no Ensino Técnico Integrado do IFFar/FW

Abstract

This paper presents a supervised machine learning model for predicting academic failure in technical education programs integrated with secondary education at the Instituto Federal Farroupilha – Frederico Westphalen Campus. The study is motivated by the role of academic failure as a factor associated with student dropout. Institutional academic data were prepared and analyzed using binary classification models implemented with the Scikit-learn library. The models were evaluated using stratified cross-validation and standard performance metrics. The results indicate that the proposed approach can support early identification of students at risk and assist academic management actions.