Table 3.

Comprehensive evaluation indicators of various machine learning algorithms

Methods	F1-score	Sensitivity	Precision	Specificity	Hamming loss	Accuracy	G-mean	AUC
ARF-OOBEE	0.9022 ± 0.0098	0.8949 ± 0.0118	0.9151 ± 0.0165	0.9805 ± 0.0338	0.0296 ± 0.0055	0.9704 ± 0.0168	0.9367 ± 0.0138	0.9830 ± 0.0202
GcForest	0.9140 ± 0.0145	0.8980 ± 0.0144	0.9420 ± 0.0169	0.9810 ± 0.0392	0.0252 ± 0.0078	0.9748 ± 0.0210	0.9386 ± 0.0236	0.9528 ± 0.0214
LR	0.8052 ± 0.0136	0.7905 ± 0.0110	0.8622 ± 0.0160	0.9581 ± 0.0300	0.0520 ± 0.0079	0.9480 ± 0.0196	0.8703 ± 0.0210	0.9616 ± 0.0225
Naive Bayes	0.7587 ± 0.0148	0.8085 ± 0.0106	0.7404 ± 0.0130	0.9113 ± 0.0380	0.0962 ± 0.038	0.9038 ± 0.0213	0.8584 ± 0.0220	0.9153 ± 0.0222
MLP	0.7673 ± 0.0152	0.7532 ± 0.0126	0.8327 ± 0.0165	0.9409 ± 0.0099	0.0745 ± 0.0380	0.9255 ± 0.0226	0.8418 ± 0.0232	0.9070 ± 0.0236
SVM	0.7411 ± 0.0133	0.7949 ± 0.0119	0.7137 ± 0.0135	0.8941 ± 0.0333	0.1090 ± 0.0083	0.8910 ± 0.0212	0.8430 ± 0.0231	0.8789 ± 0.0230
XGBoost	0.8804 ± 0.0116	0.8552 ± 0.0114	0.9435 ± 0.0176	0.9725 ± 0.0353	0.0335 ± 0.0079	0.9665 ± 0.0185	0.9120 ± 0.0227	0.9726 ± 0.0189

ARF-OOBEE, adaptive random forest-out of bag-easy ensemble; AUC, area under the curve; LR, logistic regression; MLP, multilayer perceptron; SVM, support vector machine; XGBoost, extreme gradient boosting.