Research Article
AWSMOTE: An SVM-Based Adaptive Weighted SMOTE for Class-Imbalance Learning
Table 8
G-mean with different oversampling methods from real datasets.
| Id | NO-RS | SMOTE | BLSMOTE | ADASYN | DBSMOTE | AWSMOTE |
| 1 | 0.9609 | 0.9700 | 0.9623 | 0.9626 | 0.9579 | 0.9717 | 2 | 0.8259 | 0.9117 | 0.8896 | 0.9083 | 0.9131 | 0.9173 | 3 | 0.1870 | 0.9904 | —— | 0.9953 | 0.9948 | 0.9972 | 4 | 0.9526 | 0.9640 | —— | 0.9609 | 0.9700 | 0.9709 | 5 | 0.6686 | 0.9150 | 0.9370 | 0.9475 | 0.9111 | 0.9519 | 6 | 0.8923 | 0.8970 | 0.9204 | 0.9228 | 0.8971 | 0.9270 | 7 | 0.7069 | 0.9927 | 0.9852 | 0.9930 | 0.9875 | 0.9934 | 8 | 0.2635 | 0.8393 | 0.8587 | 0.8985 | 0.8118 | 0.9001 | 9 | 0.8795 | 0.9344 | 0.9380 | 0.9305 | 0.8742 | 0.9482 | 10 | 0.9474 | 0.9531 | 0.9728 | 0.9728 | 0.9749 | 0.9763 | 11 | 0.8865 | 0.9550 | 0.9545 | 0.9522 | 0.9489 | 0.9556 | 12 | 0.8647 | 0.9930 | —— | 0.9835 | 0.9943 | 0.9959 | 13 | 0.0404 | 0.9871 | —— | 0.9859 | 0.9870 | 0.9874 | 14 | 0 | 0.9962 | —— | 0.9952 | 0.9896 | 0.9978 | 15 | 0.0282 | 0.9765 | —— | 0.9996 | 0.9988 | 1 | 16 | 0.0057 | 0.9939 | —— | 0.9937 | 0.9939 | 0.9972 | 17 | 0.0424 | 0.9987 | —— | 0.9985 | 0.9880 | 1 | 18 | 0.8459 | 0.9878 | —— | 0.9863 | 0.9776 | 0.9935 | 19 | 0.8629 | 0.9796 | —— | 0.9904 | 0.9874 | 0.9941 | 20 | 0.9523 | —— | —— | —— | 0.9926 | 0.9959 | 21 | 0 | 0.9798 | —— | 0.9774 | 0.9906 | 0.9875 | 22 | 0 | 0.9931 | —— | 0.9941 | 0.9922 | 0.9946 |
|
|