Classifying Imbalanced Data Sets by a Novel RE-Sample and Cost-Sensitive Stacked Generalization Method

Table 6

Results of other methods in terms of AUC.

Data set	NB	C4.5	KNN	Cost-NB	Ada-NB	Ada-Cost-NB	Bag- NB	Bag-Cost- NB	Stacking-log	Stacking-Cost-log
wisconsin	0.986	0.957	0.953	0.989	0.981	0.982	0.99	0.989	0.982	0.982
pima	0.819	0.751	0.662	0.819	0.804	0.808	0.817	0.819	0.819	0.819
haberman	0.64	0.615	0.53	0.64	0.667	0.625	0.638	0.647	0.636	0.63
Vehicle1	0.711	0.722	0.671	0.711	0.769	0.723	0.715	0.712	0.782	0.782
Vehicle0	0.814	0.93	0.92	0.814	0.784	0.781	0.808	0.808	0.977	0.977
segment0	0.987	0.971	0.996	0.987	0.944	0.959	0.987	0.986	0.999	0.999
yeast-0-3-5-9_vs_7-8	0.75	0.675	0.653	0.75	0.751	0.686	0.75	0.754	0.769	0.773
yeast-0-2-5-6_vs_3-7-8-9	0.832	0.687	0.782	0.832	0.79	0.808	0.84	0.842	0.847	0.82
vowel0	0.98	0.974	1	0.98	0.996	0.994	0.981	0.98	1	1
led7digit-0-2-4-5-6-7-8-9_vs_1	0.954	0.925	0.827	0.954	0.898	0.893	0.953	0.949	0.909	0.904
Glass2	0.716	0.812	0.587	0.716	0.737	0.735	0.748	0.737	0.713	0.721
cleveland-0_vs_4	0.909	0.824	0.754	0.909	0.826	0.891	0.915	0.913	0.979	0.976
glass-0-1-6_vs_5	0.934	0.989	0.825	0.934	0.975	0.905	0.983	0.985	0.9	0.99
car-good	0.976	0.493	0.648	0.976	0.968	0.971	0.976	0.976	0.653	0.977
flare-F	0.908	0.525	0.59	0.911	0.876	0.893	0.908	0.912	0.825	0.806
car-vgood	0.997	0.992	0.69	0.997	0.997	0.997	0.997	0.997	0.998	0.998
abalone-17_vs_7-8-9-10	0.775	0.613	0.624	0.775	0.817	0.73	0.773	0.774	0.752	0.726
Mean	0.864	0.791	0.748	0.864	0.858	0.846	0.869	0.869	0.855	0.875