<![CDATA[Chronic Kidney Disease (CKD) is a serious global health problem. However, information about how many people are affected by CKD in several countries is not very abundant and is sometimes not the same from one source to another. This research aims to increase accuracy in classifying CKD patients using the XGBoost ensemble learning method. The XGBoost model was drilled using the CKD dataset of 400 data records which were divided into training data and test data with a ratio of 70% used as training data and 30% as test data. Then an optimization technique is carried out, namely the parameter tuning process using a grid search method to find the best value using 5 parameters, namely n_estimators, max_ depth, learning_rate, Subsample, Colsample bytree. The evaluation results using the confusion matrix, were obtained with an accuracy level of 99.16%, precision 98.17%, recall 99.16% and f1-score 99.16%. So the XGBoost algorithm implementing parameter tuning techniques is a good classification method that is good enough to be applied in CKD and Not CKD classification.]]>
Copyrights © 2024
