1. 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
- Author
-
M. Ferrand, F. Barillet, B. Huquet, J.M. Trommenschlager, F. Faucon, M. Brochard, H. Larroque, S. Barbey, O. Leray, 149 rue de Bercy, Union nationale des coopératives d’élevage et d’insémination animale (UNCEIA), Domaine expérimental animal du Pin (SEA), Institut National de la Recherche Agronomique (INRA), Station d'Amélioration Génétique des Animaux (SAGA), 42 rue de Châteaudun, Centre National Interprofessionnel de l'Economie Laitière [Paris] (CNIEL), BP 70129, Institut Technique du Lait et des Produits Laitiers, and Agro-Systèmes Territoires Ressources Mirecourt (ASTER Mirecourt)
- Subjects
Accurate estimation ,Analytical chemistry ,mir ,Dairy industry ,Feature selection ,01 natural sciences ,Analytical Chemistry ,spectrometry ,Genetic algorithm ,Partial least squares regression ,Statistics ,genetic algorithm ,Spectroscopy ,chemistry.chemical_classification ,milk ,Process Chemistry and Technology ,010401 analytical chemistry ,0402 animal and dairy science ,Fatty acid ,mid-infrared ,04 agricultural and veterinary sciences ,pls ,[INFO.INFO-IA]Computer Science [cs]/Computer Aided Engineering ,040201 dairy & animal science ,Regression ,0104 chemical sciences ,Computer Science Applications ,chemistry ,Mid infrared spectrometry ,partial least square ,regression ,fatty acid ,Software - Abstract
International audience; 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%.
- Published
- 2011
- Full Text
- View/download PDF