Kernelization of FLDA
Han Zi-cun · Journal of Anhui University of Technology and Science · 2004
kernel method has been of wide concern in the field of machine learning recently. It allows the efficient computation of linear classification in high-dimensional feature space, instead of non-linearly separable problem in low-dimensional input space. This paper, based on the Mika's KFD, discusses how kernel methods work from input sapce to feature space in detail with mathematical derivation. Furthermore, the proposition that the project vector can be repersented by linear combination of training samples in feature space has been proved also.