Table 9.

Hyperparameters table.

Model Name	Hyperparameter Name	Hyperparameter Options
DT	criterion	‘gini’, ‘entropy’
	splitter	‘best’, ‘random’
	max_features	‘auto’, ‘sqrt’, ‘log2’, None
KNN	n_neighbors	15 31
	weights	‘uniform’, ‘distance’
	algorithm	‘ball_tree’, ‘kd_tree’
NC	shrink_threshold	0.01, 0.1, 0.2, 0.3
GNB	var_smoothing	10−7~−12
MNB	alpha	0, 0.1, 0.5, 0.8,1
CNB	alpha	0, 0.1, 0.5, 0.8,1
BNB	alpha	0, 0.1, 0.5, 0.8,1
MLR	solver	‘newton-cg’, ‘lbfgs’, ‘saga’, ‘sag’
RRC	alpha	1e-3, 1e-2, 1e-1, 1
	solver	‘svd’, ‘cholesky’, ‘lsqr’, ‘sparse_cg’, ‘sag’, ‘saga’
LCSGD	loss	‘hinge’, ‘log’, ‘modified_huber’, ‘squared_hinge’, ‘perceptron’
	alpha	1e-3, 1e-2, 1e-1, 1
	learning_rate	‘constant’, ‘optimal’, ‘invscaling’, ‘adaptive’
	eta0	0.01, 0.001, 0.0001
PAC	C	0.001, 0.01, 0.1,1
	loss	‘hinge’, ‘squared_hinge’
SVC	loss	‘hinge’, ‘squared_hinge’
	C	0.001, 0.01, 0.1, 1
RF	n_estimators	300, 500, 800
	criterion	‘gini’, ‘entropy’
	bootstrap	True, False
	max_features	‘auto’, ‘sqrt’, ‘log2’, None
ERT	n_estimators	300, 500, 800
	criterion	‘gini’, ‘entropy’
	bootstrap	True, False
	max_features	‘auto’, ‘sqrt’, ‘log2’, None
GBT	loss	deviance, exponential
	learning_rate	0.1, 0.01, 0.001, 0.1
	subsample	0.1, 0.5, 0.9
	n_estimators	300, 500, 800
	max_features	‘auto’, ‘sqrt’, ‘log2’, None
EGBT	tree_method	‘auto’, ‘exact’, ‘approx’, ‘hist’
	grow_policy	‘depthwise’, ‘lossguide’
	n_estimators	300, 500, 800
	learning_rate	0.001, 0.01
	max_depth	10, 15, 20, 50, 100