Comparative Analysis of Machine Learning Models for Student Performance Forecasting in Higher Education | IJCT Volume 12 – Issue 5 | IJCT-V12I5P79

Accurately forecasting student performance has become a critical focus in higher education, enabling institutions to identify at-risk learners and implement timely interventions to enhance academic achievement. With the increasing availability of digital learning data, machine learning techniques offer promising tools for modeling and predicting student outcomes. This study presents a comparative analysis of six prominent algorithms namely Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, Artificial Neural Network, and XGBoost to evaluate their effectiveness in forecasting student achievement using demographic, behavioral, and academic variables. Following systematic data preprocessing and hyperparameter optimization, each model’s performance was assessed using key evaluation metrics, including accuracy, precision, recall, and ROC-AUC. The results indicate that ensemble-based approaches such as Random Forest and XGBoost achieve superior predictive performance and generalization capabilities, while simpler models demonstrate efficiency and interpretability in less complex data environments. The findings contribute to the growing field of educational data mining by highlighting the potential of machine learning to support evidence-based academic planning and personalized learning interventions in higher education.

In conclusion, the comprehensive evaluation and visualization of machine learning models for student performance prediction reveal that ensemble methods such as Random Forest and XGBoost consistently deliver superior predictive accuracy and robustness compared to traditional non-ensemble approaches like Logistic Regression and Decision Tree. Through rigorous statistical testing, including paired t-tests and ANOVA, these performance gains are shown to be statistically significant and not attributable to random variation, while advanced visualizations such as boxplots, violin plots, and ROC curves provide intuitive, multidimensional insights into the distribution, stability, and discriminative power of each model. However, the study also highlights the importance of balancing predictive power with interpretability and computational efficiency, as ensemble models, despite their accuracy, can present challenges in transparency and resource demands. By integrating both quantitative metrics and effective visual communication, this research empowers educators and administrators to make informed, context-sensitive decisions about model deployment, ultimately supporting more targeted, data-driven interventions and fostering a culture of evidence-based practice in higher education analytics.

K. C. Gonugunta and K. Leo, “Role of Data-Driven Decision Making in Enhancing Higher Education Performance: A Comprehensive Analysis of Analytics in Institutional Management,” International Journal of Acta Informatica, vol. 3, no. 1, pp. 149-159, June 2024. E. Kurilovas, “On data-driven decision-making for quality education,” Computers in Human Behavior, vol. 107, p. 105774, June 2020. B. Prahani, J. Alfin, A. Fuad, H. Saphira, E. Hariyono and N. Suprapto, “Learning management system (LMS) research during 1991-2021: How technology affects education,” International Journal of Emerging Technologies in Learning (iJET), vol. 17, no. 17, pp. 28-49, September 2022. A. Rabelo, M. W. Rodrigues, C. Nobre, S. Isotani and L. Zárate, “Educational data mining and learning analytics: A review of educational management in e-learning,” Information Discovery and Delivery, vol. 52, no. 2, pp. 149-163, March 2024. S. M. Buring, A. Williams and T. Cavanaugh, “The life raft to keep students afloat: Early detection, supplemental instruction, tutoring, and self-directed remediation,” Currents in Pharmacy Teaching and Learning, vol. 14, no. 8, pp. 1060-1067, August 2022. P. Joseph-Richard, J. Uhomoibhi and A. Jaffrey, “Predictive learning analytics and the creation of emotionally adaptive learning environments in higher education institutions: a study of students’ affect responses,” The International Journal of Information and Learning Technology, vol. 38, no. 2, pp. 243-257, March 2021. H. Singh, B. Kaur, A. Sharma and A. Singh, “Framework for suggesting corrective actions to help students intended at risk of low performance based on experimental study of college students using explainable machine learning model,” Education and Information Technologies, vol. 29, no. 7, pp. 7997-8034, May 2024. S. R. Sihare, “Student Dropout Analysis in Higher Education and Retention by Artificial Intelligence and Machine Learning,” SN Computer Science, vol. 5, no. 2, p. 202, January 2024. K. Fahd, S. Venkatraman, S. J. Miah and K. Ahmed, “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, vol. 27, no. 3, pp. 3743-3775, April 2022. T. Kyriazos and M. Poga, “Application of machine learning models in social sciences: managing nonlinear relationships,” Encyclopedia, vol. 4, no. 4, pp. 1790-1805, November 2024. L. Lin, D. Zhou, J. Wang and Y. Wang, “A systematic review of big data driven education evaluation,” Sage Open, vol. 14, no. 2, April 2024. J. Fan, “A big data and neural networks driven approach to design students management system,” Soft Computing, vol. 28, no. 2, pp. 1255-1276, January 2024. A. Joshi, P. Saggar, R. Jain, M. Sharma, D. Gupta and A. Khanna, “CatBoost — An Ensemble Machine Learning Model for Prediction and Classification of Student Academic Performance,” Advances in Data Science and Adaptive Analysis, vol. 13, no. 03n04, p. 2141002, July 2021. I. Alhamad and H. P. Singh, “Predicting Dropout at Master Level Using Educational Data Mining: A Case of Public Health Students in Saudi Arabia,” Revista Amazonia Investiga, vol. 13, no. 74, pp. 264-275, February 2024.

Call for Paper (Hub) Fast Paper Publication IJCT Indexing & Databases Submit Manuscript Call for Paper: Sep–Oct 2025 Call for Paper: Publish Research