DocumentCode
3099516
Title
Fisher discriminant analysis with kernels
Author
Mika, Sebastian ; Rätsch, Gunnar ; Weston, Jason ; Schölkopf, Bernhard ; Müller, Klaus-Robert
Author_Institution
GMD FIRST, Berlin, Germany
fYear
1999
fDate
36373
Firstpage
41
Lastpage
48
Abstract
A non-linear classification technique based on Fisher´s discriminant is proposed. The main ingredient is the kernel trick which allows the efficient computation of Fisher discriminant in feature space. The linear classification in feature space corresponds to a (powerful) non-linear decision function in input space. Large scale simulations demonstrate the competitiveness of our approach
Keywords
Bayes methods; decision theory; feature extraction; learning (artificial intelligence); neural nets; pattern classification; Fisher discriminant analysis; feature space; input space; kernels; linear classification; nonlinear classification technique; nonlinear decision function; Algorithm design and analysis; Closed-form solution; Computational modeling; Feature extraction; Gaussian distribution; Kernel; Large-scale systems; Principal component analysis; Probability; Support vector machines;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks for Signal Processing IX, 1999. Proceedings of the 1999 IEEE Signal Processing Society Workshop.
Conference_Location
Madison, WI
Print_ISBN
0-7803-5673-X
Type
conf
DOI
10.1109/NNSP.1999.788121
Filename
788121
Link To Document