Recognition of Multiple Imbalanced Cancer Types Based on DNA Microarray Data Using Ensemble Classifiers

Table 5

𝐺-mean of various classification methods on eight datasets, where bold represents the best result, underline denotes the second best, and italic labels the worst one in each column, respectively.

Methods	Brain_Tumor1	Brain_Tumor2	Leukemia1	Leukemia2	Lung_Cancer	SRBCT	11_Tumors	14_Tumors
SVM-OAA	0.1012	0.6021	0.9473	0.9354	0.8362	0.9984	0.7981	0.0759
SVM-OAO	0.0279	0.6109	0.9358	0.9253	0.8417	0.9722	0.8042	0.0325
SVM-DDAG	0.1469	0.6128	0.9198	0.9074	0.8158	0.9954	0.7659	0.0468
SVM-ECOC	0.1538	0.5895	0.9436	0.9446	0.8402	0.9946	0.8125	0.0256
SVM-OAA(THR)	0.5754	0.6923	0.9426	0.9658	0.9465	0.9786	0.8143	0.1463
SVM-OAA(RUS)	0.2861	0.6052	0.9369	0.9542	0.8982	0.9994	0.8269	0.1578
EnSVM-OAA	0.0288	0.5963	0.9194	0.9403	0.8540	1.0000	0.8284	0.0886
MCSVM	0.4791	0.6281	0.9335	0.9252	0.8876	0.9688	0.8042	0.1059
Ramp-MCSVM	0.5258	0.7288	0.9517	0.9387	0.9012	0.9734	0.8548	0.1472
AdaBoost.NC	0.4326	0.6644	0.9763	0.9526	0.9349	1.0000	0.8206	0.0652
EnSVM-OAA(THR)	0.6177	0.7362	0.9727	0.9718	0.9617	0.9858	0.8562	0.1742
EnSVM-OAA(RUS)	0.4025	0.6457	0.9655	0.9588	0.9165	1.0000	0.8776	0.1983