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

Table 7

Results of other methods in terms of GeoMean.

Data set	NB	C4.5	KNN	Cost-NB	Ada-NB	Ada-Cost-NB	Bag- NB	Bag-Cost- NB	Stacking-log	Stacking-Cost-log
wisconsin	0.965	0.954	0.953	0.968	0.959	0.967	0.968	0.966	0.964	0.970
pima	0.719	0.697	0.649	0.736	0.715	0.739	0.723	0.731	0.703	0.741
haberman	0.419	0.566	0.455	0.586	0.466	0.573	0.434	0.594	0.361	0.605
Vehicle1	0.677	0.622	0.652	0.674	0.607	0.677	0.673	0.661	0.561	0.712
Vehicle0	0.724	0.912	0.919	0.726	0.735	0.733	0.719	0.733	0.894	0.946
segment0	0.897	0.977	0.995	0.892	0.897	0.906	0.902	0.895	0.994	0.993
yeast-0-3-5-9_vs_7-8	0.467	0.463	0.609	0.464	0.466	0.550	0.445	0.485	0.444	0.603
yeast-0-2-5-6_vs_3-7-8-9	0.762	0.590	0.762	0.793	0.762	0.792	0.762	0.787	0.706	0.798
vowel0	0.920	0.969	1.000	0.913	0.944	0.929	0.921	0.914	1.000	1.000
led7digit-0-2-4-5-6-7-8-9_vs_1	0.853	0.874	0.818	0.869	0.863	0.870	0.845	0.850	0.846	0.874
Glass2	0.578	0.472	0.470	0.564	0.578	0.610	0.586	0.568	0.000	0.607
cleveland-0_vs_4	0.899	0.724	0.722	0.936	0.611	0.863	0.866	0.861	0.866	0.866
glass-0-1-6_vs_5	0.932	0.877	0.810	0.932	0.932	0.872	0.872	0.869	0.814	0.877
car-good	0.000	0.000	0.558	0.794	0.520	0.783	0.000	0.778	0.561	0.813
flare-F	0.751	0.000	0.450	0.781	0.541	0.647	0.738	0.781	0.000	0.774
car-vgood	0.496	0.888	0.626	0.986	0.917	0.925	0.481	0.971	0.957	0.988
abalone-17_vs_7-8-9-10	0.628	0.321	0.506	0.640	0.319	0.368	0.627	0.630	0.293	0.664
Mean	0.687	0.641	0.703	0.779	0.696	0.753	0.680	0.769	0.645	0.814