Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 28 Results for the random forest depth 4 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0381	0.9152	0.0694	0.0848	0.9306	0.0441	0.9228	0.1376	0.0226
F-meas NC	0.1973	0.8520	0.0051	0.1480	0.9949	0.3822	0.9205	0.4561	0.2545
G-mean	0.0362	0.9189	0.0771	0.0811	0.9229	0.0398	0.9208	0.1302	0.0203
G-mean NC	0.0575	0.8955	0.0336	0.1045	0.9664	0.0885	0.9301	0.1988	0.0467
MCC	0.0362	0.9189	0.0771	0.0811	0.9229	0.0398	0.9208	0.1302	0.0203
MCC NC	0.0619	0.8931	0.0303	0.1069	0.9697	0.0975	0.9305	0.2093	0.0517
Precision	0.0382	0.9152	0.0694	0.0848	0.9306	0.0441	0.9228	0.1376	0.0226
Precision NC	0.6686	0.7846	0.0006	0.2154	0.9994	0.7638	0.8845	0.7702	0.7717
C	0.1000	0.8693	0.0142	0.1307	0.9858	0.1739	0.9255	0.2868	0.0968
D	0.5000	0.8398	0.0006	0.1602	0.9994	0.7746	0.9160	0.7771	0.7212