On the gradient-based algorithm for matrix factorization applied to dimensionality reduction

Nikulin, Vladimir and McLachlan, Geoffrey J. (2010). On the gradient-based algorithm for matrix factorization applied to dimensionality reduction. In: Ana Fred, Joaquim Filipe and Hugo Gamboa, BIOSTEC 2010: Biomedical Engineering Systems and Technologies. Proceedings of the Third International Joint Conference on Biomedical Engineering Systems and Technologies. BIOINFORMATICS 2010: 1st International Conference on Bioinformatics, Valencia, Spain, (147-152). 20-23 January 2010.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Author Nikulin, Vladimir
McLachlan, Geoffrey J.
Title of paper On the gradient-based algorithm for matrix factorization applied to dimensionality reduction
Conference name BIOINFORMATICS 2010: 1st International Conference on Bioinformatics
Conference location Valencia, Spain
Conference dates 20-23 January 2010
Convener Institute for Systems and Technologies of Information, Control and Communication
Proceedings title BIOSTEC 2010: Biomedical Engineering Systems and Technologies. Proceedings of the Third International Joint Conference on Biomedical Engineering Systems and Technologies
Journal name BIOINFORMATICS 2010 - Proceedings of the 1st International Conference on Bioinformatics
Place of Publication Portugal
Publisher Institute for Systems and Technologies of Information, Control and Communication
Publication Year 2010
Sub-type Fully published paper
ISBN 9789896740191
9896740194
Editor Ana Fred
Joaquim Filipe
Hugo Gamboa
Start page 147
End page 152
Total pages 6
Collection year 2012
Language eng
Formatted Abstract/Summary
The high dimensionality of microarray data, the expressions of thousands of genes in a much smaller number of samples, presents challenges that affect the applicability of the analytical results. In principle, it would be better to describe the data in terms of a small number of metagenes, derived as a result of matrix factorisation, which could reduce noise while still capturing the essential features of the data. We propose a fast and general method for matrix factorization which is based on decomposition by parts that can reduce the dimension of expression data from thousands of genes to several factors. Unlike classification and regression, matrix decomposition requires no response variable and thus falls into category of unsupervised learning methods. We demonstrate the effectiveness of this approach to the supervised classification of gene expression data.
Copyright © 2010 INSTICC – Institute for Systems and Technologies of Information, Control and Communication All rights reserved
Keyword Cross-validation
Gene expression data
Gradient-based optimisation
Matrix factorisation
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status UQ
Additional Notes Authors prepress title: "A gradient-based algorithm for matrix factorization applied to dimensionality reduction". BIOINFORMATICS is part of BIOSTEC - The International Joint Conference on Biomedical Engineering Systems and Technologies.

Document type: Conference Paper
Collection: School of Mathematics and Physics
 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Tue, 22 Feb 2011, 16:27:31 EST by Kay Mackie on behalf of Mathematics