Predicción del bajo rendimiento académico en programación de computadoras mediante técnicas de minería de datos educativa en muestras reducidas
Palabras clave:
Modelo de ensamble, Random forest, XGBoost, Bajo rendimiento académico, Minería de datos educativa.Resumen
El objetivo de esta investigación fue demostrar la viabilidad de un modelo predictivo de bajo rendimiento académico aplicable en contextos con datos limitados (n=94 estudiantes en su primer ciclo académico), integrando variables psicoeducativas (rasgos de personalidad y clima escolar) y técnicas especializadas (SMOTE + ReliefF + ensambles). Mediante técnicas de minería de datos educativa (MDE), este estudio se presenta como cuantitativo y no experimental aplicándose dos instrumentos de escala Likert: uno midiendo la motivación y otro evaluando los rasgos de personalidad basado en el modelo de los Cinco Grandes (Apertura, Responsabilidad, Extraversión, Amabilidad y Neurotismo). El conjunto de datos se procesó utilizando SMOTE (para equilibrar clases minoritarias) y ReliefF (para selección de características), reduciendo de 45 variables a 16 predictores claves. Se evaluaron cuatro algoritmos base (Random Forest, XGBoost, Gradient Boosting, Extra Trees) y un clasificador de ensamble por votación. Se empleó validación cruzada (k=10) y métricas como exactitud, precisión, sensibilidad, F-measure y AUC para medir el rendimiento. Sin embargo, la sensibilidad (50%) evidenció dificultades para detectar casos positivos, probablemente debido al desbalance de clases. Estos resultados demuestran la eficacia de las técnicas del método de ensamble en la predicción del bajo rendimiento académico, proporcionando a los educadores información accionable para diseñar intervenciones tempranas y mejorar, el modelo de ensamble logró la mayor exactitud (87%), precisión (71%) y AUC (72%), superando los resultados académicos.
Referencias
[1] Khan, I., Al Sadiri, A., Ahmad, A. R., & Jabeur, N. (2019). Tracking student performance in introductory programming by means of machine learning. In 2019 4th mec international conference on big data and smart city (icbdsc) (págs. 1-6). IEEE.
[2] Hasan, R., Palaniappan, S., Mahmood, S., Abbas, A., Sarker, K. U., & Sattar, M. U. (2020). Predicting student performance in higher educational institutions using video learning analytics and data mining techniques. Applied Sciences, 10(11), 3894.
[3] Franco, E. A., Martínez, R., & Domínguez, V. (2021). Modelos predictivos de riesgo académico en carreras de computación con minería de datos educativos. Revista de Educación a Distancia (RED), 21 (66).
[4] Gabatino, T., Ogawa, M. B. C., & Crosby, M. E. (2022, June). Abstracting the Understanding and Application of Cognitive Load in Computational Thinking and Modularized Learning. In International Conference on Human-Computer Interaction (pp. 273-286). Cham: Springer International Publishing.
[5] Salmon, A., Hammer, K., Santos, E. A., & Becker, B. A. (2025). Debugging Without Error Messages: How LLM Prompting Strategy Affects Programming Error Explanation Effectiveness. arXiv preprint arXiv:2501.05706.
[6] Selvakani, D., & Vasumathi, K. (2025). Enhancing Dropout Prediction in Higher Education using a Hybrid Machine Learning Approach. Journal of Computer Science, 18, 188-212.
[7] Messer, M., Brown, C., Kölling, M., & Shi, M. (2024). Automated Grading and Feedback Tools for Programming Education: A Systematic Review. ACM Transactions on Computing Education, 24 (1), 1-43.
[8] Cheng, Y.-P., Shen, P.-D., Hung, M.-L., Tsai, C.-W., Lin, C.-H., & Hsu, L. C. (2022). Applying online content-based knowledge awareness and team learning to develop students’ programming skills, reduce their anxiety, and regulate cognitive load in a cloud classroom. Universal Access in the Information Society, 21(2), 557–572.
[9] Margulieux, L. E., Morrison, B. B., & Decker, A. (2020). Reducing withdrawal and failure rates in introductory programming with subgoal labeled worked examples. International Journal of STEM Education, 7, 1-16.
[10] Navarro, R. E. (2016). El rendimiento académico: concepto, investigación y desarrollo. REICE. Revista Iberoamericana Sobre Calidad, Eficacia Y Cambio En Educación, 1(2).
[11] López-Torres, R., Pino, R., & Pedrero, E. (2018). Deserción Escolar. En R. Lopez-Torres, R. Pino, & E. Pedrero, Logro Escolar y Deserción desde el pensamiento complejo (Vol. 1, pág. 11). Académica Española.
[12] Reyes, N, Meneses, A y Díaz, A. (2022). Planificación y gestión del tiempo académico de estudiantes universitarios. Form. Univ. vol.15 no.1
[13] Suhas S Athani, Sharath A Kodli, Mayur N Banavasi, and P.G. Sunitha Hiremath. (2017). Student Performance Predictor using Multiclass Support Vector Classification Algorithm. In 2017 International Conference on Signal Processing and Communication (ICSPC). IEEE, 341–346. https://doi.org/10.1109/CSPC.2017.8305866
[14] Géron, A. (2022). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems. " O'Reilly Media, Inc.".
[15] Zaffar, M., Hashmani, M., & Savita, K. (2018). A study of prediction models for students enrolled in programming subjects. In 2018 4th International Conference on Computer and Information Sciences (ICCOINS) (págs. 1-5). IEEE.
[16] Bedregal-Alpaca, N., Aruquipa-Velazco, D., & Cornejo-Aparicio, V. (2020). Técnicas de data Mining para extraer perfiles comportamiento académico y predecir la deserción universitaria. Revista Ibérica de Sistemas e Tecnologias de Informação, (E27), 592-604
[17] Bergin, S., Mooney, A., Ghent, J., & Quille, K. (2015). Using machine learning techniques to predict introductory programming performance. International Journal of Computer Science and Software Engineering, 4, págs. 323-328.
[18] Contreras, L., Fuentes, H., y Rivas E. (2021) Análisis del rendimiento académico mediante técnicas de aprendizaje automático con métodos de ensamble revista Redipe, (10) (13)
[19] Yaacob, W. W., Sobri, N. M., Nasir, S. M., Norshahidi, N. D., & Husin, W. W. (2020). Predicting student drop-out in higher institution using data mining techniques. (I. Publishing, Ed.) Journal of Physics: Conference Series, 1496 (1), 012005.
[20] Redroban, C., Saavedra, J., Leon, M., Nuñez, S., & Echeverria, F. (2023, May). Educational Data Mining: A Predictive Model to Reduce Student Dropout. In Proceedings of International Conference on Information Technology and Applications: ICITA 2022 (pp. 713-721). Singapore: Springer Nature Singapore.
[21] Soto, C. J. & John, O. P. (2017). The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. Journal of personality and social psychology, 113(1), 117.
[22] Selvakani, D., & Vasumathi, K. (2025). Enhancing Dropout Prediction in Higher Education using a Hybrid Machine Learning Approach. Journal of Computer Science, 18, 188-212.
[23] Santoso, J. T., Ginantra, N. L. W. S. R., Arifin, M., Riinawati, R., Sudrajat, D., & Rahim, R. (2021). Comparison of classification data mining C4. 5 and Naïve Bayes algorithms of EDM dataset. TEM Journal, 10(4), 1738-1744.
[24] Vidhya, R., & Vadivu, G. (2021). Retracted article: towards developing an ensemble based two-level student classification model (ESCM) using advanced learning patterns and analytics. Journal of Ambient Intelligence and Humanized Computing, 12(7), 7095-7105.
[25] Tjahyadi, H., & Tude, K. N. (2025). The Implementation of Educational Data Mining in Predicting Students’ Academic Achievement in Mathematics at a Private Elementary School. International Journal of Information and Education Technology, 15(1).
[26] Bello, F. A., Kóhler, J., Hinrechsen, K., Araya, V., Hidalgo, L., & Jara, J. L. (2020, November). Using machine learning methods to identify significant variables for the prediction of first-year Informatics Engineering students dropout. In 2020 39th International Conference of the Chilean Computer Science Society (SCCC) (pp. 1-5). IEEE.
[27] Veerasamy, A.K.; D’Souza, D.; Apiola, M.-V.; Laakso, M.-J.; Salakoski, T. Using early assessment performance as early warning signs to identify at-risk students in programming courses. In Proceedings of the 2020 IEEE Frontiers in Education Conference (FIE), Uppsala, Sweden, 21–24 October 2020.
[28] Đambić, G.; Krajcar, M.; Bele, D. Machine learning model for early detection of higher education students that need additional attention in introductory programming courses. Int. J. Digit. Technol. Econ. 2016, 1, 1–11.
[29] Adhikari, A., Jain, L. C. & Prasad B. (2017). A state-of-the-art review of knowledge discovery in multiple databases. Journal of Intelligent Systems, 26(1), 23-24.
[30] Agrawal, P., Abutarboush, H. F., Ganesh T., & Mohamed, A. W. (2021). Metaheuristic algoritms on feature selection: A survey of one decade of research (2009-2019). Ieee Access, 9, 26766-26791.
[31] Urbanowicz, R., Meeker, M., La Cava, W., Olson, R., & Moore, J. (2018). Relief-based feature selection: Introduction and review. Journal of Biomedical Informatics (85), 189-203.
[32] Coelho, F. Castro, C. Braga, A. P., & Verleysen, M. (2019). Semi-supervised relevance index for feature selection. Neural Computing and Applications, 31, 989-997.
[33] Wang, S., Dai, Y., Shen, J., & Xuan, J. (2021). Research on expansion and classification of imbalanced data based on SMOTE algorithm. Scientific reports, 11(1), 24039.
[34] Kabari, L. G., & Onwuka, U. C. (2019). Comparison of bagging and voting ensemble machine learning algorithm as a classifier. International Journals of Advanced Research in Computer Science and software engineering, 19-23.
[35] Abdar, M., Pourpanah, F., Hussain, S., Rezazadegan, D., Liu, L., Ghavamzadeh, M., ... & Nahavandi, S. (2021). A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information fusion, 76, 243-297.
[36] Hastie, T., Tibshirani, R., & Friedman, J. (2017). The elements of statistical learning: data mining, inference, and prediction.
