Models for predicting student dropout in undergraduate courses at the Federal University of São João del-Rei
Student dropout is the abandonment of studies by formally enrolled students. This phenomenon is influenced by different factors, such as loss of interest, financial difficulties, lack of investments and appropriate policies, among others. The impacts of evasion reach different levels, from the interrupted training of the student, the waste of resources and even society, which loses a future professional. Although the analysis of evasion is a complex task, advances in computational techniques such as artificial intelligence have enabled analysis and inference from data, aiming at understanding and even predicting this phenomenon. This work aims to predict evasion in the context of higher education at the Federal University of São João del-Rei (UFSJ) considering periods before and during the COVID-19 pandemic, through the application of machine learning techniques. In addition, to evaluate the main variables related to the dropout phenomenon in both periods in search of differences. Data from graduated, dropped out and enrolled students between the years 2018 and 2021 of UFSJ undergraduate courses are included in the analysis and creation of prediction models. Sociodemographic variables, course-related information and student performance are selected as possible predictors. The initial analysis of the data made it possible to assess the quality of the available data, the frequency distribution of the variables and carry out their preparation. The database after preparation has 10657 records and 23 variables in addition to the response variable, allowing to proceed to the next steps of the methodology for creating prediction models.