Educational Data Mining in Predicting Academic Performance: a prognosis based on the curricular path taken

• Léo Manoel Lopes da Silva Garcia State University of Mato Grosso http://orcid.org/0000-0003-4861-8830
• Daiany Francisca Lara State University of Mato Grosso https://orcid.org/0000-0002-0458-9196
• Raquel Salcedo Gomes Federal University of Rio Grande do Sul https://orcid.org/0000-0001-9497-513X
• Sílvio César Cazella Federal University of Rio Grande do Sul http://orcid.org/0000-0003-2343-893X

DOI: https://doi.org/10.5753/sbie.2022.225221

Abstract

This work presents the evaluation of predictive models for the identification of students at risk of failing in specific subjects. For this purpose, the curricular path previously carried out by the student before taking a certain course is used as a predictor attribute. The impact of using load balancing techniques on predictive model evaluation metrics is investigated. The results highlighted the best performances for the Random Forest, J48 and IBK algorithms, presenting an Accuracy from 71% to 81% and Recall from 75% to 93%, reflecting a significant improvement when using the SMOTE oversampling technique for balancing charge.

Keywords: Educational data mining, academic trajectory, academic achievement

References

Akmeşe, Ö. F., Kör, H., and Erbay, H. (2021). “Use Of Machine Learning Techniques For The Forecast Of Student Achievement In Higher Education”. Information Technologies and Learning Tools, 82(2), https://doi.org/10.33407/itlt.v82i2.4178

Alturki, S., e Alturki, N. (2021). “Using educational data mining to predict students’ academic performance for applying early interventions”. Journal of Information Technology Education: Innovations in Practice, 20. https://doi.org/10.28945/4835

Anoopkumar, M. and Zubair Rahman, A. M. J. Md. (2018). “Bound Model of Clustering and Classification (BMCC) for Proficient Performance Prediction of Didactical Outcomes of Students”. International Journal of Advanced Computer Science and Applications 9(11), http://dx.doi.org/10.14569/IJACSA.2018.091133

Garcia, L. M. L. S., Lara, D. F., e Antunes, F. (2020). “Análise da Retenção no Ensino Superior: um Estudo de Caso em um Curso de Sistemas de Informação”. Revista da Faculdade de Educação 34:15-38. https://doi.org/10.30681/21787476.2020.34.1538.

Garcia, L. M. L. S.; Lara, D. F.; Gomes, R. S.; e Cazella, S. C. (2022). “The Discovery of Knowledge in Educational Databases: A Literature Review with Emphasis on Pre-processing and Postprocessing”. The Turkish Online Journal of Educational Technology (TOJET), v. 21, p. 75-87, 2022.

Manhães, L. M. B., Cruz, S. M. S. (2019) “Predição do Desempenho Acadêmico de Alunos da Graduação Utilizando Mineração de Dados”. In: XIX Simpósio de Pesquisa Operacional e Logística Marinha. Rio de Janeiro-RJ. Novembro de 2019.

Mengash, H. A. (2020).“Using Data Mining Techniques to Predict Student Performance to Support Decision Making in University Admission Systems,” in IEEE Access, vol. 8, pp. 55462-55470, doi: 10.1109/ACCESS.2020.2981905.

Miguéis, V. L., Freitas, Ana., Garcia, Paulo J.V. Silva, André. (2018). “Early segmentation of students according to their academic performance: A predictive modelling approach”. Decision Support Systems, Volume 115, Pages 36-51, ISSN 0167-9236, https://doi.org/10.1016/j.dss.2018.09.001

Pabreja, K. (2017). Comparison of Different Classification Techniques for Educational Data. IJISSS vol.9, no.1: pp.54-67. http://doi.org/10.4018/IJISSS.2017010104

Romero, C., Ventura, S. (2020). Educational data mining and learning analytics: An updated survey. WIRES Data Mining and Knowledge Discovery, 10(3), 3. https://doi.org/10.1002/widm.1355 doi:10.1002/widm.1355.

Souza, V. F., e Cazella, S. C. (2022). “Mineração De Dados Educacionais Com Algoritmos De regressão: Um Estudo Sobre a predição Do Desempenho”. Revista Educar Mais 6 :183-98. https://doi.org/10.15536/reducarmais.6.2022.2691.

Tinto, V. (2017). Through the eyes of students. Journal of College Student Retention: Research, Theory & Practice, 19(3), 254–269.