Title :
Principal tangent data reduction
Author :
Hunt, Thomas ; Krener, Arthur J.
Author_Institution :
Dept. of Math., Univ. of California, Davis, CA, USA
Abstract :
There is a need to be able to find patterns in high dimensional data sets. Often these patterns are described as lower dimensional manifolds possibly of varying dimension that more or less fit the data. We present a new algorithm for doing this. It is a form of nonlinear principle component analysis.
Keywords :
data reduction; principal component analysis; high dimensional data sets; lower dimensional manifolds; nonlinear principle component analysis; principal tangent data reduction; Automatic control; Automation; Covariance matrix; Eigenvalues and eigenfunctions; Mathematics; Piecewise linear techniques; Principal component analysis; Sun; USA Councils; Nonlinear dimension reduction; manifold learning;
Conference_Titel :
Control and Automation, 2009. ICCA 2009. IEEE International Conference on
Conference_Location :
Christchurch
Print_ISBN :
978-1-4244-4706-0
Electronic_ISBN :
978-1-4244-4707-7
DOI :
10.1109/ICCA.2009.5410162
