Low-Rank Kernel-Based Semisupervised Discriminant Analysis

Semisupervised Discriminant Analysis (SDA) aims at dimensionality reduction with both limited labeled data and copiousunlabeled data, but it may fail to discover the intrinsic geometry structure when the data manifold is highly nonlinear. The kerneltrick is widely used to map the original nonlinearly separable problem to an intrinsically larger dimensionality space where theclasses are linearly separable. Inspired by low-rank representation (LLR), we proposed a novel kernel SDA method called low-rankkernel-based SDA (LRKSDA) algorithm where the LRR is used as the kernel representation. Since LRR can capture the global datastructures and get the lowest rank representation in a parameter-free way, the low-rank kernel method is extremely effective androbust for kinds of data. Extensive experiments on public databases show that the proposed LRKSDA dimensionality reductionalgorithm can achieve better performance than other related kernel SDA methods