Premium
Positive matrix factorization applied to a curve resolution problem
Author(s) -
Xie YuLong,
Hopke Philip K.,
Paatero Pentti
Publication year - 1998
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/(sici)1099-128x(199811/12)12:6<357::aid-cem523>3.0.co;2-s
Subject(s) - factorization , matrix decomposition , matrix (chemical analysis) , resolution (logic) , non negative matrix factorization , spectral line , diffusion , chemistry , analytical chemistry (journal) , computational physics , nuclear magnetic resonance , statistical physics , physics , algorithm , mathematics , computer science , thermodynamics , chromatography , artificial intelligence , quantum mechanics , eigenvalues and eigenvectors
Positive matrix factorization (PMF) is a least squares approach to factor analysis which was originally developed for environmental data analysis and has been applied to several problems in resolving sources of environmental pollutants. PMF has been used as both a two‐way and three‐way data analysis tool. In this investigation, three‐way data arrays were used to explore the ability of PMF in curve resolution. Pulsed gradient spin echo (PGSE) nuclear magnetic resonance (NMR) data were measured for spectral mixtures where the concentrations of the compounds decay exponentially. Three‐way data arrays were constructed by packing different parts of the data from single experiments and were analyzed with three‐way PMF to obtain the NMR spectra, decay profiles and the self‐diffusion coefficients of constituents. Copyright © 1998 John Wiley & Sons, Ltd.