Table 3.

Results of the model hyperparameter tuning.

Model		Hyperparameters (lower bound-upper bound)
Logistic regression
	—a	—
Support vector machine
	Gamma	0.1 (0.1-10)
	Cost	7.1 (0.1-100)
	Degree	3 (1-5)
	Kernel	Radial basis
K-nearest neighbor
	K	1 (1-10)
Random forest
	Number of trees	230 (10-300)
	Depth	90 (10-100)
	Features	3 (1-25)
Extreme gradient boosting
	Number of trees	30 (10-300)
	Depth	16 (10-100)
	Eta	0.23 (0.01-0.4)
	Gamma	0.19 (0.01-0.2)
	Lambda	1.60 (0.1-2)
	Alpha	0.30 (0.1-2)
Bayesian additive regression trees
	Number of trees	30 (10-100)
	Depth	90 (10-100)
Artificial neural network
	Layers	2 (1-5)
	Neurons	16. 8 (64.2-32.2)
	Threshold	0.001 (0.1-0.001)

^aNot available.