Constrained PCA techniques for the identification of common factors in data

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

We present an analysis of a constrained principal components analysis network that identifies the common factors in data sets in a manner similar to principal factor analysis. This network responds to the covariance of the input data (not both variance and covariance as in PCA) and so is resistant to noise and varying levels of power on the inputs. The network naturally lends itself to the sparse coding of data, however, by enforcing this sparseness further we are able to decipher dual components in data.

Original languageEnglish
Pages (from-to)145-156
Number of pages12
JournalNeurocomputing
Volume22
Issue number1-3
DOIs
Publication statusPublished (in print/issue) - 20 Nov 1998

Keywords

  • Multiple causes
  • Principal factor analysis
  • Sparse coding
  • Unsupervised

Fingerprint

Dive into the research topics of 'Constrained PCA techniques for the identification of common factors in data'. Together they form a unique fingerprint.

Cite this