Abstract—The rapid expansion of e-learning systems has transformed modern education by generating large volumes of student interaction data, creating new opportunities for monitoring and predicting academic performance. However, accurately measuring student achievement in digital learning environments remains challenging due to high-dimensional data, complex learning behaviors, and feature redundancy. This study proposes a hybrid machine learning-based framework for forecasting student academic performance using e-learning data mining. Student achievement is measured in terms of predicting and classifying learners’ academic performance levels based on their behavioral, demographic, and academic interaction data. The proposed framework integrates data preprocessing, optimized feature selection, and predictive modeling to enhance accuracy and robustness. An optimization-driven feature selection strategy is employed to identify the most informative attributes, while a hybrid deep learning model is used to capture nonlinear and temporal learning patterns. Experimental evaluation conducted on a benchmark e-learning dataset demonstrates that the proposed approach achieves superior predictive performance compared to existing methods across standard evaluation metrics, including accuracy, precision, recall, F1-Score, and Area Under the Curve (AUC). Although the framework introduces additional computational complexity and is validated on a single dataset, the results indicate its strong potential for supporting early identification of at-risk students and enabling data-driven academic interventions in e-learning environments.
 
Keywords—educational access, process innovation, inclusive education, learning opportunities, e-learning, student academic performance, hybrid cheetah-based cat swarm optimization
 
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).