VAPNIK V.The nature of statistical learning theory[M].New York:Springer,1995:70-85.
[2]
SCHOLKOPF B,SMOLA A,MULLER K R.Nonlinear component analysis as a kernel eigenvalue problem[J].NeuralComputation,1998,10(5):1299-1319.
[3]
MIKA S,RATSCH G,WESTON J,et al.Fisher discriminant analysis with kernels[J].Proc IEEE Int l Workshop NeuralNetworks for Signal Processing IX,1999,8:41-48.
[4]
MIKA S,RATSCHG,SCHOLKOPF B,et al.Invariant feature extraction and classification in kernel spaces[C]∥Advancesin Neural Information Processing Systems.Cambridge:MITPress,1999:526-532.
[5]
SHAWE-TAYLOR J,CRISTIANINI N.Kernel methods for pattern analysis[M].Cambridge:Cambridge University Press,2004:211-229.
[6]
YANG J,FRANGI A F,YANG J Y,et al.KPCA pus LDA:a complete kernel fisher discriminant framework for featureextraction and recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(2):230-244.
[7]
XU Y,SUN B,ZHANG C Y,et al.An implemention framework for kernel methods with high-dimensional patterns[C]∥Proceedings of the Fifth International Conference on Machine Learning and Cybernetics,Dalian:IEEE Press,2006:13-16.
[8]
SCHOLKOPF B,MIKAS,KNIRSCHB,et al.Input space vs.feature space in kernel-based methods[J].IEEE Transactionson Neural Networks,1999,10(5):1000-1017.
[9]
CHEN W,ZHANG H.The condition of kernelizing an algorithm and an equivalence between kernel methods[C]∥LectureNotes In Computer Science 4477,Berlin:Springer,2007:338-345.
