DocumentCode
2199714
Title
Parallel algorithm for principal components based on Hotelling´s iterative procedure
Author
Cabaleiro, J.C. ; Carazo, J.M. ; Zapata, E.L.
Author_Institution
Dept. Electron. & Comput., Santiago Univ., Madrid, Spain
fYear
1993
fDate
27-29 Jan 1993
Firstpage
144
Lastpage
149
Abstract
In this work, we perform a parallel implementation for obtaining the principal components based on Hotelling´s (1933) iterative procedure. We also analyze a variant of this method proposed by Hotelling for accelerating the convergence of the algorithm and we present the parallel implementation of the necessary functions. Results for a SUPERNODE system, composed of sixteen T805 transputers interconnected as a hypercube, are included
Keywords
convergence of numerical methods; hypercube networks; iterative methods; parallel algorithms; statistical analysis; transputer systems; Hotelling´s iterative procedure; SUPERNODE system; T805 transputers; algorithm convergence acceleration; hypercube; parallel algorithm; principal components analysis; Acceleration; Algorithm design and analysis; Convergence; Eigenvalues and eigenfunctions; Hypercubes; Iterative algorithms; Iterative methods; Parallel algorithms; Principal component analysis; Symmetric matrices;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel and Distributed Processing, 1993. Proceedings. Euromicro Workshop on
Conference_Location
Gran Canaria
Print_ISBN
0-8186-3610-6
Type
conf
DOI
10.1109/EMPDP.1993.336409
Filename
336409
Link To Document