Table 7. Classifiers performance of DWT-based feature extraction without oversampling using SAM and HAM-A labeling.

	HAM-A label			SAM Label
	Accuracy	Recall	Precision	Accuracy	Recall	Precision
KNN	64.00%	64.61%	65.11%	55.56%	55.56%	55.89%
LDA	72.58%	72.58%	72.03%	60.62%	60.62%	59.08%
SVM	70.04%	70.04%	72.27%	64.25%	64.25%	41.28%
RF	83.93%	83.93%	84.06%	71.37%	71.37%	71.57%
AdaBoost bagging	77.42%	77.42%	77.10%	65.09%	65.09%	63.35%
Gradient bagging	77.05%	77.05%	76.74%	66.91%	66.91%	65.60%