Identification of Relevant Attributes in Evasion in Brazilian Public Higher Education

• Bruno Claudino Pereira de Brito Federal Rural University of Pernambuco
• Rafael Ferreira Leite de Mello Federal Rural University of Pernambuco https://orcid.org/0000-0003-3548-9670
• Gabriel Alves Federal Rural University of Pernambuco https://orcid.org/0000-0002-2249-7818

DOI: https://doi.org/10.5753/cbie.sbie.2020.1032

Abstract

The dropout is a frequent problem in Brazilian universities. In addition
to the loss of the student in the academic environment, it may cause inefficient university expenses since the university budget is unable to achieve its goal of training the student. Data mining techniques with the aid of algorithms can assist in an identification of the possible evasion student. Several algorithms can be used, including Random Forest. This work aims to identify, with the help of this algorithm, the relevance of personal characteristics in students in association with demographic characteristics in the dropout process. The work was carried out on data from a federal public university in conjunction with data from the Brazilian demographic census. The results achieved, evaluated between the two databases, allow to verify which characteristics are associated with the dropout students at this university.

Keywords: Educational data mining, Learning Analytics, Evasion

References

CARRANO, Davi et al. Combinando Técnicas de Mineração de Dados para Melhorar a Detecção de Indicadores de Evasão Universitária. Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação - SBIE), [S.l.], nov. 2019.


DELGADO, M. F. et al. (2014) “Do we need hundreds of classifiers to solve real world classification problems?”, J. Machine Learning Research, v. 15, n. 1, p. 3133 – 3181.


GOLDSCHMIDT, Ronaldo; PASSOS, Emmanuel. Data mining : um guia prático. 4a Reimpressão. Rio de Janeiro: Elsevier, 2005.


HAN, J., Kamber, M., and Pei, J. (2011). Data Mining: Concepts and Techniques. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 3rd edition.


INEP. Informe estatístico do MEC revela melhoria do rendimento escolar[1998?]. Disponível em: http://portal.inep.gov.br/artigo/-/asset_publisher/B4AQV9zFY7Bv/content/informe-estatistico-do-mec-revela-melhoria-do-rendimento-escolar/21206 . Acesso em: 06 de Jul. 2020.


LIAW, A.; wiener, M. Classification and regression by random forest. R News, v.2, p.18-22, 2002.


LIŢOIU, N., & Oproiu, G. C. (2018). Dropout Rate in Universities–a Critical Issue of Today Education. In The International Scientific Conference eLearning and Software for Education (Vol. 3, pp. 77-83). " Carol I" National Defence University.


PASCOAL, Tulio et al. Evasão de estudantes universitários: diagnóstico a partir de dados acadêmicos e socieconômicos. Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação - SBIE), [S.l.], p. 926, nov. 2016.


QUEIROGA, Emanuel; CECHINEL, Cristian; ARAÚJO, Ricardo. Predição de estudantes com risco de evasão em cursos técnicos a distância. Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação - SBIE), [S.l.], p. 1547, out. 2017. ISSN 2316-6533.


SANTOS, Daniel Cirne Vilas-Boas dos; FALCÃO, Taciana Pontual. Acompanhamento de alunos em ambientes virtuais de aprendizagem baseado em sistemas tutores inteligentes. Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação - SBIE), [S.l.], p. 1267, out. 2017. ISSN 2316-6533.


SILVA, Regina Maria Ferreira da. Ações afirmativas e direito fundamental à educação. Uma análise à luz das cotas raciais universitárias. Revista Jurídica da Presidência, v. 14, n. 104, out. 2012.


TAN, P.; Steinbach, M. K. V. (2009). Introdução ao data mining, Mineração de Dados. Ciência Moderna.


TIBSHIRANI, Robert; FRIEDMAN, Jerome (2008). The Elements of Statistical Learning (2nd ed.). Springer. ISBN 0-387-95284-5. Hastie, Trevor