Increasing the Training Speed of SVM, the Zoutendijk Algorithm Case

Quadratic optimization fine tuning for the Support Vector Machines learning phase

“…The Zoutendijk algorithm belongs to the methods of feasible directions (Chvatal, 2000;Ibarra-Orozco & Hernandez-Gress, 2005). This class of methods solves a nonlinear programming problem by moving from a feasible point to an improved feasible point.…”