Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 26 Results for the Extremely randomized trees depth 8 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0507	0.9213	0.0651	0.0787	0.9349	0.0470	0.9280	0.1431	0.0241
F-meas NC	0.1665	0.8628	0.0066	0.1372	0.9934	0.3168	0.9257	0.4066	0.1964
G-mean	0.0478	0.9254	0.0736	0.0746	0.9264	0.0420	0.9258	0.1347	0.0215
G-mean NC	0.0658	0.9063	0.0407	0.0937	0.9593	0.0764	0.9323	0.1840	0.0400
MCC	0.0478	0.9254	0.0737	0.0746	0.9263	0.0419	0.9257	0.1346	0.0215
MCC NC	0.0674	0.9055	0.0390	0.0945	0.9610	0.0799	0.9327	0.1883	0.0419
Precision	0.0507	0.9213	0.0651	0.0787	0.9349	0.0470	0.9280	0.1431	0.0241
Precision NC	0.3681	0.8323	0.0013	0.1677	0.9987	0.6793	0.9114	0.6963	0.6006
C	0.1000	0.8876	0.0178	0.1124	0.9822	0.1466	0.9335	0.2631	0.0800
D	0.5000	0.8167	0.0004	0.1833	0.9996	0.7911	0.9033	0.7915	0.7685