C-Support Vector Classification the Estimation of the MS Subgroups Classification with Selected Kernels and Parameters
Yeliz Karaca, Şengül Hayta, Rana Karabudak · European Journal of Pure and Applied Mathematics · 2016
The study has classified the subgroups of Multiple Sclerosis using Support Vector Machines. C-SVC algorithm, one of the SVM classifiers of multi class, has been utilized for the classification of MS subgroups. For this purpose, 120 MS patient and 19 healthy individuals have been included in our study. Through Magnetic Resonance Imaging (MRI), the number of lesion diameter and Expanded Disability Status Scale data are applied through C- Support Vector Classifier (C-SVC). By applying the data onto Radial Basis Funrtion kernel, Polynomial kernel, Sigmoid kernel and Linear kernel, four of the kernel type of C-SVC algorithm, the accuracy rates of MS subgroups classification and the computation time during the training procedure are computed and compared. Having applied C- Support Vector Classifier on MS subgroups, classification achievement of Healthy individual and MS subgroups, namely that of RRMS, SPMS end PPMS has been measured.