DocumentCode
328248
Title
Principal component analysis is a group action of SO(N) which minimizes an entropy function
Author
Takahashi, Tetsuya
Author_Institution
RIKEN, Inst. of Phys. & Chem. Res., Saitama, Japan
Volume
1
fYear
1993
fDate
25-29 Oct. 1993
Firstpage
355
Abstract
Gives a new interpretation for PCA (principal component analysis) by defining a quantity which evaluates the "goodness" of the relationship between a data set and a basis. The quantity takes the same form of entropy in Shannon\´s information theory. It is showed that PCA is equivalent to a group action such that entropy is minimized.
Keywords
matrix algebra; minimum entropy methods; neural nets; SO(N); Shannon´s information theory; entropy function; group action; principal component analysis; Chemical analysis; Eigenvalues and eigenfunctions; Entropy; Equations; Information theory; Laboratories; Neural networks; Neurons; Principal component analysis; Tin;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN
0-7803-1421-2
Type
conf
DOI
10.1109/IJCNN.1993.713930
Filename
713930
Link To Document