Theory and Modern Applications
Table 2 Results of “adult income” dataset
(ν,N,L) | Training accuracy / Test accuracy / AUC | |
|---|---|---|
Proposed method | (0.01,2,121) | 0.858 / 0.860 / 0.915 |
(0.01,5,121) | 0.858 / 0.860 / 0.915 | |
(0.01,3,363) | 0.758 / 0.788 / 0.854 | |
(0.01,30,121) | 0.858 / 0.859 / 0.915 | |
Existing methods | SVC | 0.764 / 0.764 / 0.770 |
RFC | 0.833 / 0.837 / 0.702 | |
LightGBM | 0.865 / 0.856 / 0.780 | |
XGBoost | 0.864 / 0.857 / 0.774 |