Table 3.

Performance of ML model I with different classification algorithms for training and test datasets

5-fold cross validation performance of classification algorithms on the training dataset
Algorithms	Precision	Recall	F1-score	AUC-ROC	Accuracy
SVM	86.92 ± 5.18	82.68 ± 3.94	84.64 ± 3.58	92.53 ± 3.34	85.04 ± 3.41
Extra Trees	85.77 ± 6.05	83.58 ± 7.19	84.26 ± 3.62	92.82 ± 3.64	84.52 ± 3.21
Random Forest	85.14 ± 6.11	83.45 ± 7.72	83.92 ± 4.47	92.08 ± 4.24	84.16 ± 4.04
AdaBoost	88.06 ± 7.36	80.52 ± 9.79	83.63 ± 6.36	93.69 ± 3.61	84.47 ± 5.63
XGBoost	82.99 ± 6.23	82.81 ± 7.04	83.00 ± 5.33	92.39 ± 4.58	82.74 ± 5.18
LR	82.80 ± 5.21	79.21 ± 12.17	80.65 ± 8.36	90.01± 4.06	81.55 ± 6.93
Performance of classification algorithms on the test dataset
SVM	93.75	85.71	89.55	93.89	90.14
Random Forest	88.78	79.54	83.84	91.60	84.94
Extra Trees	89.94	76.79	82.80	91.29	84.31
AdaBoost	77.42	68.57	72.72	84.84	74.65
XGBoost	83.87	74.28	78.78	88.17	80.28
LR	87.10	77.14	81.81	88.01	83.09