Weighted maximum margin discriminant analysis with kernels

A new kernel-based learning algorithm, called kernel weighted maximum margin discriminant analysis (KWMMDA), is presented in this paper. Different from the previous discriminant analysis algorithms based on the traditional Fisher discriminant criterion, KWMMDA is derived based on a new discriminant criterion, called weighted maximum margin criterion (WMMC). The better performance of KWMMDA is demonstrated by experiments on real data set.

[1]  Wenming Zheng,et al.  Locally nearest neighbor classifiers for pattern classification , 2004, Pattern Recognit..

[2]  Wenming Zheng,et al.  A Modified Algorithm for Generalized Discriminant Analysis , 2004, Neural Computation.

[3]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[4]  B. Scholkopf,et al.  Fisher discriminant analysis with kernels , 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468).

[5]  G. Baudat,et al.  Generalized Discriminant Analysis Using a Kernel Approach , 2000, Neural Computation.

[6]  Daoqiang Zhang,et al.  Efficient and robust feature extraction by maximum margin criterion , 2003, IEEE Transactions on Neural Networks.