Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 30 Results for the extremely randomized trees depth 8 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0211	0.9134	0.0612	0.0866	0.9388	0.0494	0.9259	0.1465	0.0254
F-meas NC	0.1246	0.8518	0.0025	0.1482	0.9975	0.5367	0.9217	0.5785	0.4020
G-mean	0.0196	0.9189	0.0703	0.0811	0.9297	0.0435	0.9242	0.1370	0.0223
G-mean NC	0.0315	0.8980	0.0315	0.1020	0.9685	0.0952	0.9324	0.2069	0.0504
MCC	0.0196	0.9191	0.0704	0.0809	0.9296	0.0435	0.9243	0.1370	0.0223
MCC NC	0.0349	0.8949	0.0273	0.1051	0.9727	0.1110	0.9329	0.2240	0.0596
Precision	0.0211	0.9134	0.0612	0.0866	0.9388	0.0494	0.9259	0.1465	0.0254
Precision NC	0.5029	0.7598	0.0003	0.2402	0.9997	0.7823	0.8707	0.7845	0.8149
C	0.0357	0.8910	0.0238	0.1090	0.9762	0.1146	0.9326	0.2299	0.0612
D	0.5000	0.7872	0.0003	0.2128	0.9997	0.7981	0.8869	0.7981	0.8105