<![CDATA[Air pollution due to the growth of industry and motorized vehicles seriously threatens human health. Clean air is essential, but pollutant contamination can cause acute respiratory illnesses and other illnesses. Several studies have been carried out to anticipate this air pollution. Various algorithms, methods, and data balancing techniques have been implemented, but still need to be done to obtain better accuracy results. Therefore, this study aims to classify heart disease using the XGBoost and Random Forest algorithms and implement the SMOTE technique to overcome data imbalance. This research produces a Random Forest algorithm with SMOTE implementation with splitting 80:20, which has the best accuracy with an accuracy of 92.4%, an average AUC of 0.98, and a log loss of 0.2366, which shows that SMOTE has succeeded in improving model performance in classifying minority classes. Based on the results obtained, the XGBoost and Random Forest algorithms after SMOTE are superior to the model with SMOTE, with accuracy above 90%.]]>
Copyrights © 2024
