Table 5.

Average performance evaluation of machine learning models for all test datasets.

Dataset	Model	Accuracy (%)	Precision (%)	Recall or Sensitivity (%)	Specificity (%)	F1-Score (%)	AUC (%)	MCC (%)
First Dataset [2]	LR	67.14	97.50	63.93	88.89	77.23	62.10	35.73
	KNN	67.14	92.50	64.91	76.92	76.29	62.90	32.87
	DT	78.57	85.00	79.07	77.78	81.93	77.50	55.92
	SVM	68.57	90.00	66.67	75.00	76.60	65.00	35.36
	NB	67.14	70.00	71.79	61.29	70.89	66.70	33.21
	ET	80.00	97.50	75.00	94.44	84.78	77.10	61.33
	RF	78.57	92.50	75.51	85.71	83.15	76.30	56.69
	XGBOOST	76.72	85.28	79.52	70.60	81.85	74.95	49.89
Second dataset (OSR Dataset) [6]	LR	84.73	81.99	87.82	81.82	84.80	84.80	69.66
	KNN	77.83	72.99	82.35	73.97	77.39	78.00	56.19
	DT	75.62	76.30	76.67	74.49	76.48	75.60	51.17
	SVM	83.25	81.52	85.57	80.98	83.50	83.30	66.60
	NB	79.80	83.89	78.67	81.22	81.19	79.60	59.58
	ET	84.98	82.46	87.88	82.21	85.09	85.10	70.12
	RF	83.99	81.04	87.24	80.95	84.03	84.10	68.21
	XGBOOST	81.09	81.37	80.07	82.15	80.62	81.12	62.24
Third dataset [9]	LR	88.00	33.33	38.46	92.70	35.71	63.7	29.22
	KNN	90.67	20.00	60.00	91.72	30.00	59.30	30.95
	DT	80.67	60.00	28.13	94.92	38.30	71.50	31.46
	SVM	90.67	46.67	53.85	94.16	50.00	71.10	45.02
	NB	83.33	60.00	32.14	95.08	41.86	73.00	35.36
	ET	92.00	33.33	71.43	93.01	45.45	65.90	45.30
	RF	90.67	46.67	53.85	94.16	50.00	71.10	45.02
	XGBOOST	88.64	44.96	61.19	91.77	50.03	69.92	45.70