How to build a robust model with only a few reference values: a chemometric challenge at 'Chimiométrie 2007'

J.A.F. Pierna (1) , F. Chauchard (2) , S. Preys (2) , J.M. Roger (3) , O. Galtier (4) , V. Baeten (1) , P. Dardenne (1)

1 Quality of agricultural products department
2 Ondalys
3 UMR ITAP - Information – Technologies – Analyse Environnementale – Procédés Agricoles
4 Université Paul Cézanne - Aix-Marseille 3

Following up on the success of previous chemometric challenges arranged during the annual congress organised by the French Chemometrics Society, the organisation committee decided to repeat the idea for the Chimiometrie 2007 event (http://www.chimiometrie.org/) held in Lyon, France (2930 November) by featuring another dataset on its website. As for the first contest in 2004, this dataset was selected to test the ability of participants to apply regression methods to NIR data. The aim of Challenge 2007 was to perform a calibration model as robust and precise as possible using a data set with only a few reference values available and submitted to different perturbation factors. The committee received nine answers; this paper summarizes the best three approaches, as well as the approach proposed by the organisers.

Mots clés: CHIMIOMETRIE; INFRAROUGE; SPECTROSCOPIE; ESTIMATION DES PARAMETRES; MULTIVARIATE CALIBRATION; ROBUSTNESS; EXTERNAL PARAMETER; ORTHOGONALISATION

Domaines: Ingénierie de l'environnement

J.A.F. Pierna, F. Chauchard, S. Preys, J.M. Roger, O. Galtier, et al.. How to build a robust model with only a few reference values: a chemometric challenge at 'Chimiométrie 2007'. Chemometrics and Intelligent Laboratory Systems, 2011, 106 (2), p. 152 - p. 159. ⟨10.1016/j.chemolab.2010.05.015⟩. ⟨hal-00615447⟩