DATA SCIENCE EM BASE ESCOLAR

UMA PERSPECTIVA ESTATÍSTICA DA EVASÃO ESCOLAR

Abstract

An important resource that all human beings need to rely on is education. In light of this statement, it is evident that education in Brazil faces serious problems of structure, infrastructure, and other investments, and among the cataloged issues, there is one that is more questioned and researched: school dropout. In this scenario, the general objective of this research is to establish a correlation between dropout rates in IT courses and the profile of the students, and the specific objectives include data collection for the focus database, unbiased treatment of the collected data, training of the machine learning model, and visualization and analysis of the results. The first part of the research is already underway, highlighting the historical background and data collection to provide the necessary data for the creation of the model and the required theoretical basis. This is done by adopting a methodology based on a review of specific and applied literature in the field, including the exploration of published works such as articles, monographs, theses, dissertations, as well as official databases. The final text is constructed based on consultations with academic texts and relevant documents for the discussion. The second part of the research discusses data cleaning through techniques such as ETL, with a focus on maximizing the results for our model. The application of the chosen techniques for training our machine learning model and the cataloging of the results are expected. Finally, the obtained results will be analyzed using statistical and graphical demonstrations, aiming to find the best prediction result through statistical data accompanied by graphs and observations about the project.