Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 20 Results for the random forest depth 4 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.2546	0.9352	0.0845	0.0648	0.9155	0.0377	0.9252	0.1273	0.0193
F-meas NC	0.2900	0.9256	0.0629	0.0744	0.9371	0.0519	0.9312	0.1502	0.0268
G-mean	0.2368	0.9384	0.0994	0.0616	0.9006	0.0322	0.9192	0.1167	0.0164
G-mean NC	0.2908	0.9252	0.0622	0.0748	0.9378	0.0522	0.9313	0.1508	0.0269
MCC	0.2352	0.9390	0.1008	0.0610	0.8992	0.0318	0.9188	0.1158	0.0162
MCC NC	0.3023	0.9224	0.0557	0.0776	0.9443	0.0578	0.9331	0.1592	0.0299
Precision	0.2570	0.9344	0.0824	0.0656	0.9176	0.0386	0.9258	0.1288	0.0197
Precision NC	0.5203	0.8756	0.0121	0.1244	0.9879	0.2453	0.9299	0.3463	0.1545
C	0.5000	0.8801	0.0122	0.1199	0.9878	0.2044	0.9323	0.3155	0.1164
D	0.5000	0.8801	0.0122	0.1199	0.9878	0.2044	0.9323	0.3155	0.1164