Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 36 Results for the random forest depth 4 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0036	0.9037	0.0668	0.0963	0.9332	0.0452	0.9182	0.1386	0.0232
F-meas NC	0.3479	0.8276	0.0004	0.1724	0.9996	0.8001	0.9094	0.8007	0.7763
G-mean	0.0035	0.9055	0.0714	0.0945	0.9286	0.0423	0.9169	0.1337	0.0217
G-mean NC	0.0053	0.8787	0.0336	0.1213	0.9664	0.0942	0.9212	0.2014	0.0504
MCC	0.0035	0.9055	0.0714	0.0945	0.9286	0.0423	0.9169	0.1337	0.0217
MCC NC	0.0067	0.8746	0.0274	0.1254	0.9726	0.1190	0.9220	0.2274	0.0652
Precision	0.0036	0.9037	0.0668	0.0963	0.9332	0.0452	0.9182	0.1386	0.0232
Precision NC	0.8941	0.5714	0.0001	0.4286	0.9999	0.6930	0.7541	0.7115	0.8917
C	0.0122	0.8551	0.0109	0.1449	0.9891	0.2109	0.9195	0.3182	0.1204
D	0.5000	0.7892	0.0003	0.2108	0.9997	0.8149	0.8881	0.8152	0.8431