ABSTRACT
Support vector machine is put forward by Vapnik and his partner in AT&T Bell laboratory. It is a universal learning algorithm which is based on statistic learning theory. By applying structural risk minimization principle, the support vector machine has a very good extending capacity capable of solving the problems about the samples with the characteristic of small, nonlinear, and high dimension. The core idea of the algorithm is that by using the kernel function transforms the data vectors to a feature space where the separator is linear. In fact, support vector machines algorithm is just a convex quadratic programming optimization problem, which can be solved preferably now.
