• Version
• Download 16
• File Size 509.18 KB
• File Count 1
• Create Date 19 November 2025
• Last Updated 19 November 2025

Predicting Student Academic Performance: A Machine Learning Approach for Early Intervention

 

 

Ayush Kumar
CSE

Sharda University

Greater Noida, India ayushkumarsingh23456@gmail.com

 

 

Himanshu Kumar
CSE

Sharda University

Greater Noida, India himanshu3384k@gmail.com

 

 

Jitendra Singh
CSE

Sharda University Greater Noida, India jitendravictor@gmail.com

 

 

Abstract—The increasing volume of data in educational in- stitutions provides a significant opportunity to apply machine learning for enhancing student outcomes. Early and accurate identification of students at risk of academic failure is crucial for providing timely, targeted support and improving overall retention rates. This paper presents a comprehensive comparative analysis of several supervised machine learning models for predicting student performance. We utilize a public dataset com- posed of demographic, academic, and behavioral features to train and evaluate multiple classifiers, including Logistic Regression, Decision Trees, Random Forest, Support Vector Machines, and K-Nearest Neighbors. The models are assessed based on standard performance metrics: accuracy, precision, recall, and F1-score. Our experimental results demonstrate that the Random Forest classifier achieves the highest accuracy of 89.7%, outperforming other models. We also identify key predictive features, such as previous course failures and study time, which are strong indica- tors of future performance. This study confirms the potential of machine learning models to be integrated into institutional early warning systems, enabling educators to intervene effectively and foster a more supportive learning environment.

Index Terms—Educational Data Mining, Machine Learning, Student Performance, Predictive Analytics, Early Warning Sys- tem, Classification

Download