Table 2.

Performance of each classification model in the validating dataset.

Model	AUC	Sensitivity	Accuracy	Positive Pred Value	Negative Pred Value	Recall	F1 score
Adaboost	0.898	0.7895 (0.5443, 0.9395)	0.8158 (0.6567, 0.9226)	0.8333 (0.5858, 0.9642)	0.8000 (0.5634, 0.9427)	0.6842	0.7428
LogitBoost	0.802	0.6842 (0.4345, 0.8742)	0.6842 (0.5135, 0.8250)	0.6842 (0.4345, 0.8742)	0.6842 (0.4345, 0.8742)	0.6842	0.6842
XGBoost	0.927	0.7895 (0.5443, 0.9395)	0.8421 (0.6875, 0.9398)	0.8824 (0.6356, 0.9854)	0.8095 (0.5809, 0.9455)	0.7894	0.8333
LR	0.934	0.7895 (0.5443, 0.9395)	0.8421 (0.6875, 0.9398)	0.8824 (0.6356, 0.9854)	0.8095 (0.5809, 0.9455)	0.7894	0.8333
RF	0.909	0.7895 (0.5443, 0.9395)	0.8158 (0.6567, 0.9226)	0.8333 (0.5858, 0.9642)	0.8000 (0.5634, 0.9427)	0.7894	0.8108
SVM	0.95	0.7895 (0.5443, 0.9395)	0.8421 (0.6875, 0.9398)	0.8824 (0.6356, 0.9854)	0.8095 (0.5809, 0.9455)	0.7894	0.8333
NN	0.953	0.7895 (0.5443, 0.9395)	0.8421 (0.6875, 0.9398)	0.8824 (0.6356, 0.9854)	0.8095 (0.5809, 0.9455)	0.7894	0.8333
KNN	0.945	0.8421 (0.6042, 0.9662)	0.8684 (0.7191, 0.9559)	0.8889 (0.6529, 0.9862)	0.8500 (0.6211, 0.9679)	0.8421	0.8648
DT C5.0	0.88	0.7895 (0.5443, 0.9395)	0.7368 (0.5690, 0.8660)	0.7143 (0.4782, 0.8872)	0.7647 (0.5010, 0.9319)	0.7894	0.75
NB	0.956	0.8947 (0.6686, 0.9870)	0.8421 (0.6875, 0.9398)	0.8095 (0.5809, 0.9455)	0.8824 (0.6356, 0.9854)	0.8947	0.85
GBM	0.9	0.7895 (0.5443, 0.9395)	0.8158 (0.6567, 0.9226)	0.8333 (0.5858, 0.9642)	0.8000 (0.5634, 0.9427)	0.7894	0.8108
MLP	0.917	0.7895 (0.5443, 0.9395)	0.8421 (0.6875, 0.9398)	0.8824 (0.6356, 0.9854)	0.8095 (0.5809, 0.9455)	0.7894	0.8333

AUC, area under the curve; LR, logistic regression; RF, random forest; SVM, support vector machine; NN, neural network; KNN, k-nearest neighbors; DT, decision tree; NB, naive Bayes; GBM, gradient boosting machine; MLP, multilayer perceptron.