From Regression to Classification in Support Vector Machines
Massimiliano Pontil, Ryan M. Rifkin, Theodoros Evgeniou · DSpace@MIT (Massachusetts Institute of Technology) · 1998
Abstract. We study the relation between support vector machines (SVMs) for regression (SVMR) and SVM for classification (SVMC). We show that for a given SVMC solution there exists a SVMR solution which is equivalent for a certain choice of the parameters. In particular our result is that for ɛ sufficiently close to one, the optimal hyperplane and threshold for the SVMC problem with regularization parameter Cc are equal to 1 1−ɛ times the optimal hyperplane and threshold for SVMR with regularization parameter Cr =(1−ɛ)Cc. A direct consequence of this result is that SVMC can be seen as a special case of SVMR. 1.