Dataset | M | n | G-ECOC | G-OAA | Error-ECOC | Error-OAA |
---|
| | |
η
= 5, 10, 15
| | |
GCM RM | 11 | 10 | 926 | 1260 | 0.1852 | 0.1852 |
GCM | 14 | 20 | 1314 | 423 | 0.4782 | 0.3043 |
- The performance of OAA and ECOC classifiers of size at most ⌈η·log2M⌉ on benchmark microarray datasets under bounded optimum S 2N gene selection and a public train-test partition. M and n respectively denote the number of binary classifiers used by OAA and ECOC classifiers. G-OAA and G-ECOC respectively denote the overall number of genes selected by OAA and ECOC classifiers. Error-OAA and Error-ECOC respectively denote the classification error attained by OAA and ECOC classifiers.