Premium
Nonlinear vertex discriminant analysis with reproducing kernels
Author(s) -
Wu Tong Tong,
Wu Yichao
Publication year - 2012
Publication title -
statistical analysis and data mining: the asa data science journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.381
H-Index - 33
eISSN - 1932-1872
pISSN - 1932-1864
DOI - 10.1002/sam.11137
Subject(s) - linear discriminant analysis , kernel fisher discriminant analysis , nonlinear system , pattern recognition (psychology) , kernel (algebra) , optimal discriminant analysis , artificial intelligence , discriminant , mathematics , vertex (graph theory) , multiple discriminant analysis , computer science , kernel method , algorithm , machine learning , support vector machine , combinatorics , graph , physics , quantum mechanics , facial recognition system
The novel supervised learning method of vertex discriminant analysis (VDA) has been demonstrated for its good performance in multicategory classification. The current paper explores an elaboration of VDA for nonlinear discrimination. By incorporating reproducing kernels, VDA can be generalized from linear discrimination to nonlinear discrimination. Our numerical experiments show that the new reproducing kernel‐based method leads to accurate classification for both linear and nonlinear cases. © 2012 Wiley Periodicals, Inc. Statistical Analysis and Data Mining, 2012