Multiclass classification machine based on the analytical center

Considering that for the "one versus all"(OvA) approach, repeating construction of all classifier leads to daunting computation and low efficiency of classification and multiclass classifier based on SVM, which corresponds to a simple quadratic optimization, is not very effective when the version space is asymmetric or elongated. Those problems are addressed by proposing a multiclass classifier based on the analytical center of version space, which is called M-ACM. Experiments on wine recognition and glass identification dataset demonstrate that M-ACM is validated.