Data set 1:
The data originates from a digest of an autocatalytic material. The 155 lines refer to the most prominent, least interferred, lines of those elements chosen.

The data set contains:
X: 49 × 155 (Training data set including 155 of the most prominent emission lines for Pt, Pd, Rh, Al, Mg, Ce, Zr and a few other elements)
Xtst: 10 ×155 (The corresponding data of an independent test set)
pt: 49 × 1 (Training set concentrations for Pt)
pttst: 10 × 1 (Pt concentrations of the independent test set)


Data set 2:
This data set comprises fusion data for the following elements: Au, Ag, Pt, Pd, Ir, Rh and Ru. The atomic emission lines were chosen on the basis of their minimal interference characteristics and their intensity relative to the nearest interfering peak. The original samples were autocatalytic digests and were subjected to a peroxide fusion to release all elements of interest from the matrix. The elemental concentrations were derived using a more accurate process.

The data set contains:
X: 111 × 2268 (111 samples taken from past assay work x 2268 of the most prominent ICP-AES lines).
Y: 111 × 7 (111 samples x the 7 elements: Au, Ag, Pt, Pd, Ir, Rh and Ru).

Get the data, filename: ICPAES.ZIP, available in MATLAB 7.4 format (39 + 370 kb)



Please make reference to the data by this publication:
Griffiths, M.L.; Svozil, D.; Worsfold, P.; Denham, S.; Evans, E.H. Variable reduction algorithm for atomic emission spectra: application to multivariate calibration and quantitative analysis of industrial samples, J. Anal. At. Spectrom., 2002, 17, 800-812.