Skip to main content
. 2023 Feb 16;14:1126418. doi: 10.3389/fmicb.2023.1126418

Table 2.

Performance comparison across MaxEnt, XGBoost, and ensemble models.

Model MaxEnt XGBoost Ensemble
Dataset Train Test Train Test Train Test
AUCROC 0.833 ± 0.003 0.821 ± 0.009 0.933 ± 0.002 0.854 ± 0.007 0.904 ± 0.002 0.845 ± 0.008
Sensitivity 0.622 ± 0.006 0.606 ± 0.026 0.916 ± 0.004 0.742 ± 0.18 0.781 ± 0.005 0.684 ± 0.020
Specificity 0.874 ± 0.003 0.874 ± 0.003 0.816 ± 0.003 0.816 ± 0.003 0.848 ± 0.003 0.848 ± 0.003
F1-score 0.479 ± 0.007 0.312 ± 0.008 0.548 ± 0.005 0.293 ± 0.005 0.527 ± 0.005 0.308 ± 0.005

Bolded value indicates the best performance per evaluation metric (AUCROC, Sensitivity, Specificity, and F1-score) per train or test dataset across the three modeling methods.