Table 5. Classifiers performance of PSD-based feature extraction without oversampling using SAM and HAM-A labeling.

	HAM-A label			SAM Label
	Accuracy	Recall	Precision	Accuracy	Recall	Precision
KNN	64.37%	64.37%	67.74%	58.57%	58.57%	58.11%
LDA	68.35%	68.35%	67.74%	60.39%	60.39%	58.00%
SVM	67.15%	67.15%	72.46%	64.25%	64.25%	59.63%
RF	78.74%	78.74%	78.67%	68.36%	68.36%	67.47%
AdaBoost bagging	74.52%	74.52%	74.20%	67.03%	67.03%	65.60%
Gradient Bagging	73.06%	73.06%	72.66%	64.13%	64.13%	62.44%