. 2022 Nov 21;12:19999. doi: 10.1038/s41598-022-22011-8

Table 6.

Experimental results of ML classifiers with SMOTE.

Classifier	Class	Accuracy	PPV	Sensitivity	F1 score
DT	$β$ -Thalassaemia non-carrier	0.92	0.93	0.92	0.93
	$β$ -Thalassaemia carrier		0.92	0.92	0.92
	Weighted average		0.92	0.92	0.92
GBM	$β$ -Thalassaemia non-carrier	0.92	0.92	0.92	0.92
	$β$ -Thalassaemia carrier		0.91	0.91	0.91
	Weighted average		0.92	0.92	0.92
ADA	$β$ -Thalassaemia non-carrier	0.94	0.96	0.93	0.95
	$β$ -Thalassaemia carrier		0.92	0.96	0.94
	Weighted average		0.94	0.95	0.94
SVC	$β$ -Thalassaemia non-carrier	0.92	0.96	0.89	0.92
	$β$ -Thalassaemia carrier		0.89	0.96	0.92
	Weighted average		0.92	0.92	0.92
RF	$β$ -Thalassaemia non-carrier	0.95	0.97	0.94	0.96
	$β$ -Thalassaemia carrier		0.93	0.97	0.95
	Weighted average		0.95	0.95	0.95
ETC	$β$ -Thalassaemia non-carrier	0.95	0.97	0.93	0.95
	$β$ -Thalassaemia carrier		0.93	0.97	0.95
	Weighted average		0.95	0.95	0.95
LR	$β$ -Thalassaemia non-carrier	0.93	0.96	0.91	0.93
	$β$ -Thalassaemia carrier		0.90	0.96	0.93
	Weighted average		0.93	0.93	0.93