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

Table 3

Accuracy 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	0.8596	0.6840	0.9618	0.9334	0.9515	0.9992	0.8932	0.5177
SVM-OAO	0.8611	0.6600	0.9570	0.9369	0.9388	0.9763	0.8851	0.4962
SVM-DDAG	0.8427	0.6760	0.9416	0.9278	0.8987	0.9981	0.8643	0.4865
SVM-ECOC	0.8529	0.6660	0.9558	0.9543	0.9516	0.9916	0.8915	0.5098
SVM-OAA(THR)	0.7291	0.7120	0.9158	0.9621	0.9227	0.9752	0.8862	0.5426
SVM-OAA(RUS)	0.8674	0.7320	0.9596	0.9578	0.9429	0.9988	0.8916	0.5334
EnSVM-OAA	0.8755	0.6980	0.9713	0.9459	0.9571	1.0000	0.9021	0.5638
MCSVM	0.8223	0.6460	0.9351	0.9286	0.9315	0.9628	0.8437	0.4988
Ramp-MCSVM	0.8477	0.7420	0.9417	0.9338	0.9296	0.9687	0.9146	0.5012
AdaBoost.NC	0.8516	0.6820	0.9822	0.9515	0.9597	1.0000	0.8759	0.4928
EnSVM-OAA(THR)	0.7961	0.7260	0.9634	0.9726	0.9532	0.9902	0.9017	0.6246
EnSVM-OAA(RUS)	0.8837	0.7620	0.9806	0.9604	0.9611	1.0000	0.9224	0.5974