Table 9 Part-D mean and standard deviation of AUC with varying levels of RUS (10 iterations of fivefold cross-validation)
From: Evaluating classifier performance with highly imbalanced Big Data
Class ratio | LGB | CB-16 | ET | XGB-24 | RF-GPU-32 |
|---|---|---|---|---|---|
1:1 | 0.76230 | 0.77169 | 0.97625 | 0.95292 | 0.93523 |
(0.00067) | (0.00075) | (0.00033) | (0.00059) | (0.00046) | |
1:3 | 0.76326 | 0.77125 | 0.97586 | 0.96809 | 0.95465 |
(0.00060) | (0.00074) | (0.00038) | (0.00045) | (0.00024) | |
1:9 | 0.76317 | 0.76571 | 0.97464 | 0.97256 | 0.96476 |
(0.00055) | (0.00080) | (0.00038) | (0.00037) | (0.00040) | |
1:27 | 0.76279 | 0.75888 | 0.97244 | 0.97363 | 0.96868 |
(0.00060) | (0.00062) | (0.00037) | (0.00037) | (0.00030) | |
1:81 | 0.76212 | 0.75329 | 0.96966 | 0.97348 | 0.96963 |
(0.00064) | (0.00089) | (0.00035) | (0.00039) | (0.00037) | |
Unchanged | 0.76085 | 0.74847 | 0.96746 | 0.97273 | 0.96996 |
(0.00081) | (0.00076) | (0.00038) | (0.00042) | (0.00031) |