FINDING THE OPTIMAL SEPARATING HYPERPLANE BASED ON VICINAL RISK MINIMIZATION |
| 2 | |
| 2013 |
| scientific article | 519.7 | ||
| 171-176 | vicinal risk minimization, separating hyperplane, support vector machine, feature space. |
| We investigate the methodology of support-vector machines (SVM) based on the vicinal risk minimization principle. The problem of finding the optimal separating hyperplane in the case of linearly inseparable data is considered. An algorithm of the SVM weak field is presented in the form of a convex approximation of linear risk minimization with a spherical Gaussian assessment. The methodology of constructing hyperplanes in a multidimensional space is presented. |
 |