How we should choose a kernel function in support vector machines (SVMs), is an important but difficult problem. In this paper, we discuss the properties of the solution of the M-SVMS, a variation of SVMs in two extreme cases: All feature vectors are almost orthogonal and all feature vectors are almost the same. In the former case, the solution of the z^-SVM is nearly the center of gravity of the examples given while the solution is approximated to that of the υ-SVM with the linear kernel in the latter case.
展开▼