Um método de referência com aprendizado de máquina para previsão de evasão de alunos em instituições de ensino superior particular

Abstract

University dropout certainly afflicts several higher-level private academic institutions, and the constant search for phenomena that cause it has been the focus of much research in education and educational technologies globally. Thus, predicting evasion in teaching units, in particular, becomes essential, focusing on formulating actions that reduce its impact on institutions and students in their educational development. Machine learning (ML) techniques can help teaching institutions and other educational areas predict dropout. The general objective of this work is to present a proposal for a reference model for predicting dropout using a machine learning algorithm with the decision tree technique. A hybrid research methodology is adopted with a literature review of an exploratory, explanatory, and descriptive nature of the dropout scenario with a specific methodological itinerary and research universe and collected data being the specific university environment. The objective is to create a model for analyzing university dropouts using the decision tree technique