Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 34 Results for the extremely randomized trees depth 8 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0083	0.9096	0.0607	0.0904	0.9393	0.0500	0.9241	0.1469	0.0257
F-meas NC	0.1187	0.8405	0.0008	0.1595	0.9992	0.7396	0.9163	0.7453	0.6637
G-mean	0.0076	0.9136	0.0692	0.0864	0.9308	0.0438	0.9220	0.1370	0.0224
G-mean NC	0.0155	0.8870	0.0245	0.1130	0.9755	0.1209	0.9300	0.2340	0.0654
MCC	0.0076	0.9136	0.0693	0.0864	0.9307	0.0437	0.9220	0.1370	0.0224
MCC NC	0.0169	0.8844	0.0214	0.1156	0.9786	0.1352	0.9301	0.2487	0.0739
Precision	0.0083	0.9096	0.0607	0.0904	0.9393	0.0500	0.9241	0.1469	0.0257
Precision NC	0.6452	0.6098	0.0001	0.3902	0.9999	0.7219	0.7791	0.7368	0.8960
C	0.0122	0.8947	0.0321	0.1053	0.9679	0.0881	0.9305	0.1994	0.0464
D	0.5000	0.7206	0.0002	0.2794	0.9998	0.7903	0.8482	0.7945	0.8780