“…We choose six estimator values of 50, 100, 200, 300, 400, and 500, five minimum child weight values of 1, 2, 3, 4, and 5, and five regularization alpha values of 1e-5, 1e-3, 1e-2, 0.1, and 1. The search ranges for learning rate, maximum tree depth, subsample and colsample_bytree were [0.05,0.2], ( 3 , 11 ), [0.5,1] and [0.5,1], respectively. We carefully tuned the parameters of the XGBoost algorithm to obtain the best performance, the final XGBoost model parameter settings are described in Table 3 .…”