Fluorescence data – four PARAFAC component system

Problem

The dataset was produced by Dorrit Baunsgaard for fundamental studies on PARAFAC. She has reported her findings in an internal report.

 

Get the data

The data is available in zipped MATLAB 6 or higher format. Download the data (Data) and write load data in MATLAB. If you use the data we would appreciate that you report the results to us as a courtesy of the work involved in producing and preparing the data. Also you may want to refer to the data by referring to

Jordi Riu, Bro, Jack-knife estimation of standard errors and outlier detection in PARAFAC models. Chemometrics and Intelligent Laboratory Systems, 2003, 65(1), 35-49

 

Data

27 synthetic samples containing different concentrations of four analytes (hydroquinone, tryptophan, phenylalanine and dopa) were measured in a Perkin-Elmer LS50 B fluorescence spectrometer. Each fluorescence landscape corresponding to each sample consists of 233 emission wavelengths (250-482 nm) and 24 excitation wavelengths (200-315 nm taken each 5 nm). An example of a fluorescence landscape can be seen here:

The fluorescence data is three-way. Ideally, the data is trilinear, the components of the modes corresponding to concentrations (27), emission spectra (233) and excitation spectra (24). Hence a three-way PARAFAC model should be capable of uniquely and meaningfully describing the variation in the data set.