A Novel Selective Ensemble Algorithm for Imbalanced Data Classification Based on Exploratory Undersampling

Table 4

Performance results of all methods based on 𝐺-mean evaluation metric. The values with boldface mean the best result.

Dataset	CART	RUSB	SMOB	UNBag	SMBag	AdaC	RAMO	RotF	Easy	RotE-un	RotEasy
Spambase	0.9128	0.8914	0.9391	0.9374	0.9330	0.9478	0.9474	0.9503	0.9565	0.9586	0.9457
Vote	0.9443	0.9536	0.9497	0.9524	0.9471	0.9472	0.9422	0.9457	0.9573	0.9528	0.9666
Wdbc	0.9207	0.9471	0.9648	0.9402	0.9444	0.9486	0.9663	0.9567	0.9621	0.9577	0.9688
Ionosphere	0.8621	0.9039	0.9382	0.8989	0.8942	0.9126	0.9369	0.9327	0.9215	0.9238	0.9538
Pima	0.6599	0.7446	0.7366	0.7375	0.6898	0.2401	0.7216	0.7206	0.7423	0.7253	0.8307
German	0.6374	0.7258	0.6683	0.7085	0.6325	0.2408	0.6527	0.6436	0.7238	0.7203	0.8007
Phoneme	0.8363	0.7836	0.9079	0.8826	0.8854	0.8864	0.9054	0.8875	0.8966	0.8917	0.9222
Haberman	0.5299	0.5932	0.5975	0.6311	0.4673	0.2395	0.4876	0.4746	0.6167	0.6332	0.7452
Vehicle	0.6585	0.7981	0.7602	0.7828	0.6626	0.2243	0.6845	0.6542	0.7944	0.7352	0.8379
Cmc	0.5314	0.6308	0.5793	0.6625	0.5406	0.2389	0.5544	0.5121	0.6519	0.6569	0.7610
House	0.5539	0.7196	0.7394	0.7393	0.6172	0.4818	0.6725	0.5386	0.7789	0.7369	0.7874
Scrapie	0.3980	0.5663	0.3968	0.6017	0.5651	0.1860	0.4667	0.3951	0.5969	0.5926	0.6471
Yeast	0.8144	0.9089	0.8956	0.9300	0.8706	0.8740	0.8705	0.8218	0.9289	0.9285	0.9314
Mfeat_zer	0.8167	0.9268	0.9645	0.9373	0.8789	0.8839	0.9342	0.8894	0.9621	0.9638	0.9568
Mfeat_kar	0.8282	0.9182	0.9638	0.9075	0.8754	0.8730	0.9177	0.8638	0.9554	0.9579	0.9586
Satimage	0.7218	0.8375	0.8633	0.8797	0.7849	0.7645	0.8067	0.7423	0.8915	0.8895	0.9032
Abalone7	0.4591	0.5625	0.5733	0.7845	0.4991	0.3692	0.4504	0.2616	0.7911	0.7921	0.7939
Sick	0.9064	0.9417	0.9458	0.9599	0.9322	0.9342	0.9330	0.9285	0.9588	0.9566	0.9642
Cbands	0.8531	0.9371	0.9631	0.9557	0.9138	0.9150	0.9390	0.9049	0.9663	0.9744	0.9763
Ozone	0.3627	0.2427	0.6495	0.8101	0.4911	0.2280	0.2649	0.1143	0.8121	0.8161	0.8407
Average	0.7103	0.7765	0.7998	0.8320	0.7513	0.6168	0.7527	0.7069	0.8432	0.8382	0.8746