“…In the testing part, the XGBoost model had the best prediction results with respect to R2 , MAE, RMSE, MARE, NSE and RSR (i.e., R 2 = 0.955, MAE = 59.929, RMSE = 80.653, MARE = 6.6, NSE = 0.950, and RSR = 0.225) compared to AdaBoost (i.e., R 2 = 0.950, MAE = 70.383, RMSE = 90.665, MARE = 8.252, NSE = 0.936, and RSR = 0.253), RF (i.e., R 2 = 0.945, MAE = 69.030, RMSE = 86.348, MARE = 8.014, NSE = 0.942, and RSR = 0.241), DT (i.e., R 2 = 0.0.925, MAE = 74.450, RMSE = 99.822, MARE = 8.775, NSE = 0.923, and RSR = 0.278) and SVM (i.e., R 2 = 0.878, MAE = 98.320, RMSE = 128.027, MARE = 10.991, NSE = 0.873, and RSR = 0.357) as shown in Table5.…”