SVM works by mapping data to a high-dimensional feature space in order to categorize data points that are otherwise not linearly separable. A separator between the categories is discovered, and the data are processed so that the separator may be drawn as a hyperplane.
Vectors of Support Support vectors are data points that are closer to the hyperplane and have an effect on its location and orientation. Using these support vectors, we maximize the classifier's margin. The position of the hyperplane will change if the support vectors are removed.