Abstract
A nonlinear supervised feature extraction algorithm that directly combines Fisher's criterion function with a preliminary non linear projection of vectors in pattern space will be described. After some computational details are given, a comparison with Fisher's linear method will be made over a concrete example.
Cite
CITATION STYLE
APA
Cruz, C. S., & Dorronsoro, J. (1996). A nonlinear discriminant algorithm for data projection and feature extraction. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1112 LNCS, pp. 563–568). Springer Verlag. https://doi.org/10.1007/3-540-61510-5_96
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
Already have an account? Sign in
Sign up for free