<![CDATA[Education in Indonesia currently faces several challenges, particularly the inequality of educational facilities in rural areas, leading to lower academic achievement compared to urban students. This study differs from previous research that focused solely on machine learning with academic data. Using a data-driven predictive analytics approach, the research aims to analyze factors influencing student academic performance, such as study hours, sleep hours, previous scores, and extracurricular involvement. Several machine learning algorithms including Linear Regression, Support Vector Regression (SVR), Random Forest, K-Nearest Neighbors (KNN), and XGBoost were employed to build the prediction model. The results indicated a significant correlation of 0.92 between previous scores and academic performance. Among the five algorithms, the XGBoost model demonstrated superior performance compared to the others. This highlights the effectiveness of the XGBoost model in predicting factors that affect students' academic performance and its potential as a tool for educators to develop more effective learning strategies, ultimately aiming to enhance students' academic achievements significantly.]]>
Copyrights © 2024
