Efficient Implementation of Nonparallel Hyperplanes Classifier

Abstract

In this paper, we proposed a novel nonparallel hyper planes classifier for binary classification, termed as NHC. Though this method can be in fact proved equivalent to an improved twin support vector machine (TWSVM), it has the incomparable advantages than existing TWSVMs. First, the optimization problems in NHC can be solved efficiently by successive over relaxation (SOR) without needing to compute the large inverse matrices before training as TWSVMs usually do, Second, kernel trick can be applied directly to NHC, which is superior to existing TWSVMs. Experimental results on lots of data sets show the efficiency of our method in both computation time and classification accuracy.