Optimización de la predicción de retención estudiantil en la educación superior: un análisis comparativo de modelos de aprendizaje automático supervisados y no supervisado
DOI:
https://doi.org/10.66778/CTI.v02n01.03Palabras clave:
Retención estudiantil, educación superior, modelos predictivos, machine learning, deserciónResumen
Esta investigación aborda el problema de la deserción estudiantil en la educación superior en El Salvador, utilizando modelos de aprendizaje automático. Para ello, se empleó un conjunto de datos históricos proveniente de una universidad privada, que incluye más de 83,000 registros de estudiantes entre los años 2019 y 2024. En la fase de análisis exploratorio, se aplicaron técnicas no supervisadas como Análisis de Componentes Principales (PCA), K-means y agrupamiento jerárquico, con el objetivo de segmentar a los estudiantes e identificar perfiles con características demográficas, académicas y económicas similares. Posteriormente, se implementaron modelos predictivos supervisados, entre ellos SVM, KNN, Random Forest, AdaBoost y XGBoost, evaluando su rendimiento mediante métricas como precisión y AUC-ROC. Los resultados indicaron que Random Forest ofrece una mayor capacidad predictiva, mientras que las técnicas de clusterización permitieron identificar grupos de riesgo sin necesidad de etiquetas previas. Se observó que los estudiantes en modalidad presencial, con bajo rendimiento académico y autofinanciamiento, presentan una mayor probabilidad de abandono. La investigación concluye que la combinación de análisis exploratorio y modelos predictivos facilita el diseño de estrategias efectivas para mejorar la retención estudiantil. Además, se destaca el valor del uso institucional de estas herramientas para desarrollar intervenciones proactivas, especialmente dirigidas a estudiantes de primer año o con condiciones académicas y económicas vulnerables.
Descargas
Referencias
Abdullah, N., Mohamad, A., & Bakar, A. (2019). Predicting student dropout using clustering techniques. International Journal of Advanced Computer Science and Applications, 10(9), 73-81.
Arqawi, S., Zitawi, E., Rabaya, A., Abunasser, B., & Abu-Naser, S. (2022). Predicting University Student Retention using Artificial Intelligence. International Journal of Advanced Computer Science and Applications. https://doi. org/10.14569/ijacsa.2022.0130937
Asif, R., Merceron, A., Ali, S. A., & Haider, N.G. (2017). Analyzing undergraduate students’ performance using educational data mining. Computers & Education, 113, 177-194.
Astin, A. (1999). Student involvement: A developmental theory for higher education. Journal of College Student Development, 40(5), 518-529.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324
Cabrera, A., Nora, A., & Castaneda, M. (2006). College persistence: Structural equations modeling test of an integrated model of student retention. The Journal of Higher Education, 64(2), 123-139.
Cabrera, A., Nora, A., & Castaneda, M. (2018). The role of finances in the persistence process: A structural model. Research in Higher Education, 33(5), 571-593.
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, 785-794. https://doi. org/10.1145/2939672.2939785
Chicco, D., & Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics, 21(6), 1-13. https://doi.org/10.1186/s12864-019-6413-7
Cortez, P., & Silva, A. (2008). Using data mining to predict secondary school student performance. In Proceedings of the 5th Annual International Conference on Educational Data Mining, 5-12.
Fahd, K., Venkatraman, S., Miah, S. J., & Ahmed, K. (2022). Application of machine learning in higher education to assess student academic performance, at-risk, and attrition: A meta-analysis of literature. Education and Information Technologies, 1-33.
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27(8), 861-874.
Fernández-Delgado, M., Cernadas, E., Barro, S., & Amorim, D. (2014). ¿Do we need hundreds of classifiers to solve real world classification problems? Journal of Machine Learning Research, 15(1), 3133- 3181. Obtenido de http://jmlr.org/papers/v15/delgado14a.html
Gupta, S., Lehmann, D., & Stuart, J. (2006). Valuing customers. Journal of Marketing Research, 43(1), 7-18.
Hagedorn, L. S. (2021). How to define retention: A new look at an old problem. Journal of College Student Retention: Research, Theory & Practice, 23(1), 1-15.
Herzog, S. (2006). Estimating student retention and degree-completion time: Decision trees and neural networks vis-à-vis regression. New Directions for Institutional Research(131), 17-33.
Kotler, P., & Armstrong, G. (2018). Principles of Marketing (17th ed.). Education Pearson. Kotler, P., & Keller, K. (2016). Marketing Management (15th ed.). Pearson Education.
Kurniawan, T., & Taufiq, M. (2019). Clustering students based on their academic data using K-Means algorithm. Journal of Information
Systems Engineering and Business Intelligence, 5(2), 110-118. Obtenido de https://www.ijstr.org/final-print/aug2019/Clustering-Student-Data-Based-On-K-means-Algorithms.pdf
Martínez, R., Caballero, A., & Figueroa, R. (2021). Early detection of student dropout in higher education using machine learning techniques. Education and Information Technologies, 26, 2133-2152. Obtenido de https://www.researchgate.net/publication/372493356_ Predicting_Student_Dropout
Meneses, J., Rodríguez-Gómez, D., & Fernández, C. (2020). Determinants of higher education drop-out in Latin America: A systematic review. Journal of Latin American Studies, 52(3), 645-665.
Peng, L., Chaoli, S., Guochen, Z., & Yaochu, J. (2020). Multi-surrogate multi-tasking optimization of expensive problems. Knowledge-Based Systems. https://doi.org/10.1016/j.knosys.2020.106262
Powers, D. M. (2011). Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation. Journal of Machine Learning Technologies, 2(1), 37-63.
Qingtao, X., Xin, Z., & Chenghua, Z. (2019). Application Research of KNN Algorithm Based on Clustering in Big Data Talent Demand Information Classification. International Journal of Pattern Recognition and Artificial Intelligence, 34(19). https://doi.org/10.1142/ S0218001420500159
Romero, C., Ventura, S., & García, E. (2013). Data mining in course management systems: Moodle case study and tutorial. Computers & Education, 51(1), 368-384.
Schapire, R., & Freund, Y. (2012). Boosting: Foundations and Algorithms. MIT Press (MA). https://doi.org/10.7551/mitpress/8291.001.0001
Simões, A., & Soares, M. (2019). A model of student attrition and retention in higher. Educational Research and Evaluation, 25(2-3), 25(2-3), 133-152.
Thomas, L. (2018). Understanding student retention and engagement: Contexts and strategies for improvement. Higher Education Quarterly, 72(3), 278-289.
Tianqi , C., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Association for Computing Machinery(10), 785–794. https://doi.org/10.1145/2939672.2939785
Tinto, V. (2012). Completing college: Rethinking institutional action. University of Chicago Press. Tinto, V. (2017). Through the eyes of students. Journal of College Student Retention: Research, Theory & Practice, 19(3), 254-269.
Vandamme, J.-P., Meskens, N., & Superby, J.-F. (2017). Predicting academic performance by data mining methods. Education Economics, 15(4), 405-419. Obtenido de https://ideas.repec.org/a/taf/edecon/v15y2007i4p405-419. html
Descargas
Publicado
Número
Sección
Licencia
Derechos de autor 2026 Franklin Ivan Argueta Bermudez, Wilber Alexander Ortiz Cortez, Miguel García Torres

Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial 4.0.