[Retracted] Prediction and Analysis of Length of Stay Based on Nonlinear Weighted XGBoost Algorithm in Hospital

Table 2

Parameters setting for different classifiers.

The adopted classifiers	The setting of the predefined parameters
The proposed method	Number of trees: [10, 30, 50, 100]
	Maximum depth of tree: [3, 5, 8, 10]
	Minimum leaf node weight sum: [1, 3, 6, 9]
	Learning rate parameters: [0.05, 0.1, 0.15, 0.2]
Naive Bayes	Weight control parameters: [0.5, 1, 1.5, 2, 2.5]
XGBoost	Default parameters
	Smoothing parameters
SVM	Kernel function: RBF
	Penalty coefficient: [0.01, 0.1, 1, 10]
	Kernel parameter: [0.01, 0.001, 0.0001]
KNN	Number of nearest neighbors: [3, 5, 8, 10]
	Maximum number of leaves: [5, 8, 10, 30]
Decision tree	Number of trees: [10, 30, 50, 100]
	Maximum depth range: [3, 5, 8, 10]
	Learning rate range: [0.05, 0.1, 0.15, 0.2]