Determination of fatty acid profile in cow's milk using mid-infrared spectrometry: Interest of applying a variable selection by genetic algorithms before a PLS regression

被引：45

作者：

Ferrand, M. ^{[1
]}

Huquet, B. ^{[1
]}

Barbey, S. ^{[2
]}

Barillet, F. ^{[3
]}

Faucon, F. ^{[1
,4
]}

Larroque, H. ^{[3
]}

Leray, O. ^{[5
]}

Trommenschlager, J. M. ^{[6
]}

Brochard, M. ^{[1
]}

机构：

[1] Inst Elevage, F-75595 Paris 12, France

[2] UE INRA Pin Au Haras, F-61310 Le Pin Au Haras, France

[3] INRA SAGA, F-31326 Castanet Tolosan, France

[4] CNIEL, F-75314 Paris 09, France

[5] ACTILAIT 39, F-39802 Poligny, France

[6] UR ASTER INRA Mirecourt, F-88500 Mirecourt, France

来源：

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS | 2011年 / 106卷 / 02期

关键词：

Mid-infrared (MIR) spectrometry; Milk; Fatty acid; Genetic algorithms; Partial Least Squares (PLS) regression; LEAST-SQUARES REGRESSION; SPECTRAL DATA; BOVINE-MILK; SPECTROSCOPY; OPTIMIZATION; WAVELENGTHS; PREDICTION;

D O I：

10.1016/j.chemolab.2010.05.004

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The new challenges of the dairy industry require an accurate estimation of fine milk composition. The mid-infrared (MIR) spectrometry method appears to be a good, fast and cheap method for assessing milk fatty acid profile. Although partial least squares (PLS) regression is a very useful and powerful method to determine fine milk composition from the spectra, the estimations are not always very accurate and stable over time. Therefore a genetic algorithm (GA) combined with a PLS regression was used to produce models with a reduced number of wavelengths and a better accuracy. The results are a little sensitive to the choice of parameters in the algorithm. The number of wavelengths to consider is reduced substantially by 4 and accuracy is increased on average by 15%. (C) 2010 Elsevier B.V. All rights reserved.

引用

页码：183 / 189

页数：7

共 28 条

[1]

[Anonymous], 2009, MATL VERS 7 8 0 347

[2]

[Anonymous], 2004, Wiley InterScience electronic collection.

[3] Genetic algorithm-based method for selecting wavelengths and model size for use with partial least-squares regression: Application to near-infrared spectroscopy [J].