BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Treatment of Noise in Multivariate Data Analysis Techniques.
DTSTART:20110527T101500
DTSTAMP:20260928T185959Z
UID:0e64ccd19445b99eb156118baa74cd2cebb281dd7b827a496d53ae76
CATEGORIES:Conferences - Seminars
DESCRIPTION:Pr. S. Narasimhan\, Chemical Engineering Workshop\, Indian Ins
 titute of Technology Madras\, India.\nThe last decade has seen an explosio
 n in the quantum of data available on different systems. This has led to a
  growth in the development and use of techniques for mining this data for 
 extracting valuable information. The spectrum of applications include spee
 ch and image processing\, biomedical signal processing\, bioinformatics\, 
 envirometrics and chemometrics. Several multivariate data analysis techniq
 ues such as Principal Components Analysis (PCA) and its variants\, Non-neg
 ative Matrix Factorization (NMF)\, Independent Components Analysis (ICA) e
 tc. are among the popular techniques being currently used. Despite the fac
 t that the data obtained in many of the above applications contain a signi
 ficant amount of noise\, relatively less effort has been directed at treat
 ing noise in a systematic and theoretically rigorous manner. Methods such 
 as NMF and ICA are developed from a deterministic viewpoint\, and typicall
 y PCA is used as a pre-processing technique for dealing with noise. The pu
 rpose of this talk is to first review the conditions under which PCA is an
  optimal technique for denoising data. The Iterative PCA (IPCA) method\, w
 hich we have developed for dealing with heteroscedastic errors in a rigoro
 us manner\, by estimating both the noise parameters and the regression mod
 el simultaneously\, is also discussed. IPCA is further integrated with fun
 ctional PCA for developing a powerful combined univariate-multivariate den
 oising technique. The use of the proposed approach in developing more accu
 rate multivariate calibration models and as a pre-processing technique for
  accurate extraction of source signals from mixtures is illustrated.
LOCATION:MEC2405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR