Predicting Student Academic Performance Using Machine Learning

The introduction of the Internet of Things (IoT), Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), and Big Data have paved the way for research focused on improving the student learning experience and help to address challenges faced by the education system. Machine Learning technology analyzes data to recognize patterns and use them to make predictions. This paper introduces a ML model that classify and predict student academic success by utilizing supervised ML algorithms like Random Forest, Support Vector Machines, Gradient boosting, Decision Tree, Logistic Regression, Regression, Extreme Gradient Boosting (XGBoost), and Deep Learning. This paper aims to predict student's academic success based on historical data and identify the key factors that affect student academic success. Thus, the proposed approach offers a solution to predict student academic performance efficiently and accurately by comparing several ML models to the Deep Learning model. Results show that the Extreme Gradient Boosting (XGBoost) can predict student academic performance with an accuracy of 97.12%. Furthermore, results showed significant social and demographic features that affect student academic success. This study concludes that applying Machine Learning technology in the classroom will help educators identify gaps in student learning and enable early detection of underperforming students, thus empowering educators with informed decision-making.