Model Prediksi Kelulusan Siswa Menggunakan Algoritma Adaptive Boosting (Adaboost) dan Extreme Gradient Boosting (Xgboost)

Student graduation is one indicator in evaluating the success of the learning process, so an accurate prediction model is needed to support academic decision making. This study aims to compare the performance of prediction models generated by the AdaBoost and XGBoost ensemble learning algorithms in predicting student graduation. The stages of the study include data collection, data pre-processing, validation, evaluation, and model comparison. Model evaluation used a confusion matrix, producing measurement metrics of accuracy, precision, recall, and F1-score. The comparison results showed that both models performed very well. The AdaBoost model produced an accuracy of 0.95, precision of 0.95, recall of 0.99, and an F1-score of 0.97. Meanwhile, the XGBoost model produced an accuracy of 0.94, precision of 0.94, recall of 0.99, and an F1-score of 0.97. Based on the comparison results, AdaBoost showed slightly superior performance in terms of accuracy and precision. The results of the study show that the boosting algorithm is effective for implementation in predicting student graduation and can be used to support decision-making in the academic field.