Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 32 Results for the random forest depth 4 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0114	0.9096	0.0657	0.0904	0.9343	0.0461	0.9217	0.1406	0.0237
F-meas NC	0.3419	0.8392	0.0008	0.1608	0.9992	0.7330	0.9156	0.7413	0.6627
G-mean	0.0111	0.9116	0.0714	0.0884	0.9286	0.0424	0.9199	0.1345	0.0217
G-mean NC	0.0158	0.8886	0.0378	0.1114	0.9622	0.0802	0.9245	0.1870	0.0422
MCC	0.0111	0.9116	0.0714	0.0884	0.9286	0.0424	0.9199	0.1345	0.0217
MCC NC	0.0193	0.8823	0.0310	0.1177	0.9690	0.1007	0.9245	0.2097	0.0541
Precision	0.0114	0.9096	0.0657	0.0904	0.9343	0.0461	0.9217	0.1406	0.0237
Precision NC	0.7582	0.7147	0.0002	0.2853	0.9998	0.7689	0.8435	0.7739	0.8448
C	0.0357	0.8632	0.0125	0.1368	0.9875	0.1911	0.9231	0.3018	0.1075
D	0.5000	0.8218	0.0004	0.1782	0.9996	0.8049	0.9062	0.8049	0.7895