Student Grade Prediction Using Machine Learning

Abstract

In recent years, predictive analytics has become an essential part of higher education institutions to improve academic decision-making and student performance assessment. Machine learning techniques play a key role in predicting students’ final grades by analyzing various educational and behavioral factors. This study develops a predictive framework using the publicly available Kaggle Student-Mat dataset to forecast student grades with increased accuracy and reliability. Three machine learning algorithms—Linear Regression, Support Vector Machine (SVM), Random Forest, and K-Nearest Neighbors (KNN)—are implemented and compared based on their predictive performance. The dataset undergoes preprocessing, normalization, and feature selection to ensure model robustness. The experimental results show that Linear Regression outperforms the other models in accuracy and generalization, with Linear Regression and KNN following closely. The findings demonstrate that machine learning-based predictive models can provide valuable insights for educators and institutions to identify at-risk students early and improve educational outcomes.