A New Semi-Supervised Multi-Surface Proximal Support Vector Machine Model

The currently proposed Multi-surface Proximal Support Vector Machine Classification via Generalized Eigenvalues (GEPSVM) is an effective method on 2-class problem, which only needs to proximally solve two not parallel planes corresponding to each of two data sets, and the planes can be easily obtained by solving generalized eigenvalues. However, the not parallel planes may not be accurately determined when the labeled samples is not sufficient. To overcome this disadvantage, in this paper, we introduce a new semi-supervised multi-surface proximal support machine model, which can effectively utilize the labeled and unlabeled samples by incorporating the manifold regularization strategy. Based on this model, we propose a linear semi-supervised multi-surface classification algorithm called SGEPSVM. Further, we develop a non-linear classifier by using kernel trick called SKGEPSVM. Experiments on 8 benchmark data sets show the effectiveness of our algorithms.