Predicting Progress: An In-Depth Study of Students’ Academic Trajectories through Data Mining

Abstract

Accurately predicting students' academic trajectories is crucial for effective educational interventions. This research introduces a comprehensive predictive model for understanding and forecasting students' progress, leveraging advanced data mining techniques. Specifically, three powerful classifiers—Random Forest, Support Vector Machines (SVM), and Artificial Neural Network (ANN) are employed to explore the intricate dynamics of students' academic journeys. The study places significant emphasis on uncovering patterns related to student absence days, lecturer involvement, and punctuality within the e-learning management system. This study examines the influence of Random Forest, SVM, and ANN on students' educational achievement by utilizing Random Forest, SVM, and ANN. The model proposed here incorporating these classifiers, demonstrates a substantial improvement of up to 12% to 18% in accuracy compared to models lacking these influential features. This research contributes to the field by showcasing the effectiveness of Random Forest, SVM, and ANN in predicting academic trajectories, thereby facilitating targeted interventions and personalized strategies for student success. The findings underscore the importance of leveraging diverse classifiers to comprehensively capture the multifaceted aspects of students' academic progress.