Predicting and Enhancing Student Performance Through Data Analysis and Engagement Metrics

This research explores the use of data-driven strategies to enhance student performance and engagement in educational settings. By analyzing historical student performance data, including past grades, attendance, and extracurricular involvement, the study develops predictive models to forecast final exam outcomes and identify at-risk students. It also monitors student engagement by examining interactions with coursework and assignments to pinpoint areas of struggle. The research employs machine learning techniques, such as linear regression, random forest, and XGBoost, to create accurate predictions and identify key performance trends. Findings highlight the effectiveness of these models in improving educational strategies, allowing for targeted interventions and resource allocation. The study underscores the importance of incorporating comprehensive data and addressing academic integrity issues to maximize predictive accuracy and support student success. Among the four machine learning models—Random Forest, XGBoost, Decision Tree, and Linear Regression—the Random Forest model delivers the best performance, achieving the highest R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.921. It also outperforms the others with the lowest MAE (0.0606) and MSE (0.0066), indicating superior predictive accuracy.