A fast iterative nearest point algorithm for support vector machine classifier design

In this paper we give a new fast iterative algorithm for support vector mac hine (SVM) classifier design. The basic problem treated is one that does no t allow classification violations. The problem is converted to a problem of computing the nearest point between two convex polytopes. The suitability of two classical nearest point algorithms, due to Gilbert, and Mitchell ct at, is studied. Ideas from both these algorithms are combined and modified to derive our fast algorithm, For problems which require classification vio lations to be allowed, the violations are quadratically penalized and an id ea due to Cortes and Vapnik and FrieB is used to convert it to a problem in which there are no classification violations, Comparative computational ev aluation of our algorithm against powerful SVM methods such as Platt's sequ ential minimal optimization shows that our algorithm is very competitive.