Evasão de alunos de graduação: estudo com bancos de dados nos centros da UFES em Alegre

• Autor(a) principal: Silveira, Sergio Rodrigues
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal do Espírito Santo (riUfes)
• Texto Completo: http://repositorio.ufes.br/handle/10/11349
• topic: Permanência; Administração Pública; Evasão universitária; Mineração de dados (Computação); Classificação; 35

Evasion is a phenomenon that represents a problem for educational institutions, whether public or private, causing loss of resources for all those involved in the teaching process. Educational Data Mining is a recent area of research that has shown promise in the development of methods for the exploration of data produced in educational environments, helping in the elaboration of indicators to mitigate the factors that prevent the permanence of students. The main objective of the present research was to obtain information that provides the acquisition of knowledge related to student evasion at the Federal University of Espírito Santo and that make it possible to make decisions that favor the permanence of its students. The research carried out a quantitative approach, using secondary data to obtain patterns and models that allowed to perform predictive analyzes with its variables. As for its purposes, it was a descriptive and methodological research. As data source, tables containing social and academic information of students enrolled in the years 2007 to 2018 were used in the two UFES centers, located in the city of Alegre. Data preprocessing, transformation and mining were performed using the RapidMiner Studio program, version 9.2.001, WIN64 platform. With the obtained results, it was concluded that it is possible to increase the knowledge and the decision-making capacity of the UFES university management to reduce the rates of evasion in the institution. As a technical product resulting from the research, a set of instructions as elaborated so that the UFES public management can use the same process applied in this work to make predictions with future students