Number of the records: 1
On prior information in principal component analysis
- 1.0410628 - UTIA-B 20010097 RIV IE eng C - Conference Paper (international conference)
Šmídl, Václav - Kárný, Miroslav - Quinn, A.
On prior information in principal component analysis.
Maynooth: NUI Maynooth, 2001. In: Irish Signals and Systems Conference 2001. Proceedings. - (Shorten, R.; Ward, T.; Lysaght, T.), s. 129-134
[Irish Signals and Systems Conference 2001. Maynooth (IE), 25.06.2001-27.06.2001]
R&D Projects: GA ČR GA102/99/1564
Institutional research plan: AV0Z1075907
Keywords : PCA * prior information * dynamic medical imaging
Subject RIV: BB - Applied Statistics, Operational Research
http://library.utia.cas.cz/separaty/historie/karny-on prior information in principal component analysis.pdf
Principal component analysis is well developed and understood method of multivariate data processing. Performance of PCA depends on the amount and characteristics of the noise in the observed data. In this paper we show how the use of a Bazesian approach, and especially prior information, improves its performance.
Permanent Link: http://hdl.handle.net/11104/0130717
Number of the records: 1