Student Performance Prediction and Early Warning System

It is not a good idea to treat the prediction as isolated data. Additionally, these elements can impact the interpretation of results in models. Instead of relying only on an expected risk assessment, attention should also be given to the student's needs. A machine-learning model is employed in the study to predict how students will perform and identify students who may need early academic preparation. Using Decision Tree, Random Forest and SVM, experimental techniques were applied to the UCI Student Performance dataset. A precision of 70.89% was achieved by Random Forest and SVM. The Decision Tree had the highest recall, standing at 53.85%, with an F1-score of 51.85%. This demonstrates the importance of using multiple indicators to make an informed decision. The percentage of students at high risk is a better predictor for the Decision Tree than Random Forest and SVM. The findings show that machine learning predictions can provide useful information for identifying students who may require early academic support. However, the prediction should be considered as a decision-support signal rather than as the only basis for intervention. Teachers are obligated to consider the complete needs of an individual student and provide suitable assistance. Lack of support for missing students is a significant issue in an early intervention setting. It is suggested that future studies should concentrate on larger and more diverse datasets, such as temporal and behavioral data, class imbalance in prediction models, and connections to structured academic interventions. Decision-support signals can therefore be an important component of machine learning predictions when they are considered together with appropriate academic support.

[1]  E. Ahmed, “Student Performance Prediction Using Machine Learning Algorithms,” Applied Computational Intelligence and Soft Computing, vol. 2024, Article ID 4067721, 2024. doi: 10.1155/2024/4067721. [2]  Y. A. Alsariera, Y. Baashar, G. Alkawsi, A. S. Mustafa, A. A. Alkahtani, and N. Ali, “Assessment Algorithms for Predicting Student Performance,” Computational Intelligence and Neuroscience, vol. 2022, Article ID 4151487, 2022. doi: 10.1155/2022/4151487.  [3]  I. El Guabassi, Z. Bousalem, R. Marah, and A. Qazdar, “Comparative Analysis of Supervised Machine Learning Algorithms to Build a Predictive Model for Evaluating Students’ Performance,” International Journal of Online and Biomedical Engineering (iJOE), vol. 17, no. 2, pp. 90–105, 2021. doi: 10.3991/IJOE.V17I02.20025. [4] C.-H. Chen, S. J. H. Yang, J.-X. Weng, H. Ogata, and C.-Y. Su, “Predicting at-risk university students based on their e-book reading behaviours by using machine learning classifiers,” Australasian Journal of Educational Technology, vol. 37, no. 4, pp. 130–144, 2021. doi: 10.14742/AJET.6116.  [5] L. Yang and Z. Bai, “Study on Score Prediction Model with High Efficiency Based on Deep Learning,” Electronics, vol. 11, no. 23, Article 3995, 2022. doi: 10.3390/electronics11233995. [6]  M. Kannan and S. A. A. Raj, “Enhancing Student Performance Prediction Through Machine Learning and Deep Learning Models,” International Journal of Basic and Applied Sciences, vol. 14, no. 2, pp. 296–310, 2025. doi: 10.14419/88yz9489.  [7]  B. Albreiki, T. Habuza, and N. M. Zaki, “Extracting topological features to identify at risk students using machine learning and graph convolutional network models,” International Journal of Educational Technology in Higher Education, vol. 20, Article 23, 2023. doi: 10.1186/s41239-023- 00389-3. [8]  C. Herodotou, B. Rienties, M. Hlosta, A. Boroowa, C. Mangafa, and Z. Drahhal, “The scalable implementation of predictive learning analytics at a distance learning university: Insights from a longitudinal case study,” The Internet and Higher Education, vol. 45, Article 100725, 2020. doi: 10.1016/j.iheduc.2020.100725.  [9] Y. S. Balcioğlu and M. Artar, “Predicting academic performance of students with machine learning,” Information Development, vol. 41, no. 3, pp. 896–915, 2025. doi: 10.1177/02666669231213023. [10]  C. R. Çırak, H. Akıllı, and Y. Ekinci, “Development of an early warning system for higher education institutions by predicting first-year student academic performance,” Higher Education Quarterly, vol. 78, Article e12539, 2024. doi: 10.1111/hequ.12539.