PREDICTIVE MODELS FOR STUDENT PERFORMANCE: A MACHINE LEARNING PERSPECTIVE

Machine learning techniques are increasingly being applied across various domains, with education emerging
as a key area of focus. The application of these methods in educational research is growing rapidly, enabling the
discovery of underlying patterns in student performance. This research aims to develop a predictive model for
academic outcomes using several machine learning classification techniques, such as K-Nearest Neighbor,
Decision Tree, Support Vector Machines, Random Forest, and Gradient Boosting. The model considers various
factors, including living environment, parent relationships, educational background, employment status,
backlogs, attendance, internet availability, and smartphone usage. The goal is to predict student performance in
final exams and estimate their final grades. Such a model allows educational institutions and teachers to detect
students who may be at risk, enabling them to take early actions to improve academic performance and
enhance exam results.
Keywords: Educational Data Mining, Machine Learning, Classification, Student Academic Performance.