Recursive least squares projection twin support vector machines for nonlinear classification

During the last few years, multiple surface classification (MSC) algorithms, such as projection twin support vector machine (PTSVM), and least squares PTSVM (LSPTSVM), have attracted much attention. However, there are not any modifications of them that have been presented to handle nonlinear classification. This motivates the rush towards new classifiers. In this paper, we formulate a nonlinear version of the recently proposed LSPTSVM for binary nonlinear classification by introducing nonlinear kernel into LSPTSVM. This formulation leads to a novel nonlinear algorithm, called nonlinear LSPTSVM (NLSPTSVM). Additionally, in order to promote its generalization capability, we also extend the recursive leaning method, used for further boosting the performance of PTSVM and LSPTSVM, to the nonlinear case. Experimental results on synthetic datasets, UCI datasets and NDC datasets show that NLSPTSVM has better classification capability.