Machine learning assessment of zoonotic potential in avian influenza viruses using PB2 segment

BMC Genomics

Table 4 Classification performances

Method	Accuracy	Weighted Cohen’s Kappa	Macro F1	Weighted F1
Linear regression	0.965 ± 0.002	0.944 ± 0.004	0.856 ± 0.011	0.800 ± 0.016
LASSO regression	0.976 ± 0.004	0.974 ± 0.004	0.811 ± 0.017	0.724 ± 0.025
Ridge regression	0.984 ± 0.001	0.983 ± 0.002	0.875 ± 0.013	0.817 ± 0.018
KNN regression	0.986 ± 0.002	0.984 ± 0.002	0.887 ± 0.015	0.835 ± 0.022
MLP	0.987 ± 0.001	0.986 ± 0.001	0.894 ± 0.014	0.845 ± 0.021
Stacked Bi-LSTM w/ FCN	0.925 ± 0.015	0.920 ± 0.017	0.728 ± 0.018	0.617 ± 0.026
Random Forest	0.989 ± 0.001	0.988 ± 0.001	0.918 ± 0.008	0.880 ± 0.012
CatBoost	0.987 ± 0.001	0.986 ± 0.002	0.897 ± 0.012	0.850 ± 0.017
TabNet	0.960 ± 0.003	0.968 ± 0.002	0.837 ± 0.007	0.766 ± 0.010

ISSN: 1471-2164