AWSMOTE: An SVM-Based Adaptive Weighted SMOTE for Class-Imbalance Learning

Table 6

AUC with different oversampling methods from real datasets.

Id	NO-RS	SMOTE	BLSMOTE	ADASYN	DBSMOTE	AWSMOTE
1	0.9616	0.9742	0.9628	0.9627	0.9586	0.9771
2	0.8325	0.9152	0.8916	0.9121	0.9139	0.9185
3	0.5391	0.9904	——	0.9953	0.9948	0.9972
4	0.9533	0.9648	——	0.9619	0.9710	0.9715
5	0.7205	0.9154	0.9372	0.9483	0.9131	0.9525
6	0.8938	0.9008	0.9215	0.9235	0.8981	0.9288
7	0.7535	0.9928	0.9853	0.9930	0.9876	0.9934
8	0.5758	0.8466	0.8660	0.9022	0.8284	0.9032
9	0.8834	0.9357	0.9387	0.9315	0.8786	0.9493
10	0.9482	0.9533	0.9731	0.9732	0.9752	0.9763
11	0.8910	0.9553	0.9544	0.9525	0.9493	0.9559
12	0.8778	0.9930	——	0.9837	0.9942	0.9944
13	0.5116	0.9872	——	0.9860	0.9870	0.9874
14	0.5000	0.9963	——	0.9952	0.9896	0.9978
15	0.5100	0.9769	——	0.9995	0.9988	1
16	0.5016	0.9939	——	0.9937	0.9939	0.9942
17	0.5150	0.9987	——	0.9985	0.9882	1
18	0.8662	0.9878	——	0.9866	0.9776	0.9934
19	0.8775	0.9798	——	0.9905	0.9875	0.9941
20	0.5000	——	——	——	0.9927	0.9960
21	0.5000	0.9799	——	0.9769	0.9906	0.9875
22	0.5000	0.9931	——	0.9942	0.9923	0.9946