Optimization of Hyperparameters Introgression Algorithm for Prediction of Student Academic Performance

Abstract

Students' academic achievement is measured by test scores, knowledge, and skills gained from formal education. The importance of identifying potential academic failures motivates this research to find out the factors that affect student academics. This study aims to predict student achievement based on several factors in the internal scope and exam results by using random forest regression, decision tree, and Gradient Boosting methods. The results show that the Ensemble model dominates, with high R-squared values indicating its ability to explain variations in student academic performance and low average MAE, MSE, and RMSE values indicating better performance. The results of the model identify factors that affect variations in student performance based on the tested dataset. This research provides insights for teachers and other stakeholders to improve education by better understanding the factors that influence student academic performance.