Second derivative of interactance spectra (731–926 nm) of intact peaches and Brix values of extracted juice were used to develop a least squares support vector machine (LS-SVM) regression (based on an RBF kernel) and a PLS regression model. An iterative approach was taken with the LS-SVM regression, involving a grid search with application of a gradient-based optimisation method using a validation set for the tuning of hyperparameters, followed by pruning of the LS-SVM model with the optimised hyperparameters. The grid search approach led to five-fold faster and better determination of hyperparameters. Less than 45% of the initial 1430 calibration samples were kept in the models. In prediction of an independent test set with 120 samples, the pruned LS-SVM models performed better than the PLS model (RMSEP decreased by 9% to 14%).
CogdillR. and DardenneP., “Least-squares support vector machines for chemometrics: An introduction and evaluation”, J. Near Infrared Spectrosc.12, 93 (2004). doi: 10.1255/jnirs.412
4.
ChauchardF.CogdillR.RousselS.RogerJ.M. and Bellon-MaurelV., “Application of LS-SVM to non-linear phenomena in NIR spectroscopy: Development of a robust and portable sensor for acidity prediction in grapes”, Chemometr. Intell. Lab. Syst.16, 141 (2004). doi: 10.1016/j.chemolab.2004.01.003
5.
CogdillR.SchimleckL.JonesP.PeterG.DanielsR. and ClarkA.III, “Estimation of the physical wood properties of Pinus taeda L. radial strips using least squares support vector machines”, J. Near Infrared Spectrosc.12, 263 (2004). doi: 10.1255/jnirs.434
6.
WuD.HeY.FengS. and SunD., “Study on infrared spectroscopy technique for fast measurement of protein content in milk powder based on LS-SVM”, J. Food Eng.84, 124 (2008). doi: 10.1016/j.jfoodeng.2007.04.031
7.
SunT.LinH.XuH. and YingY., “Effect of fruit moving speed on predicting soluble solids content of ‘Cuiguan’ pears (Pomaceae pyrifolia Nakai cv. Cuiguan) using PLS and LS-SVM regression”, Postharvest Biol. Technol.51, 86–90 (2009). doi: 10.1016/j.postharvbio.2008.06.003
8.
ÜstünB.MelssenW.OudenhuijzenM. and BuydensL., “Determination of optimal support vector regression parameters by genetic algorithms and simplex optimisation”, Anal. Chim. Acta544, 292 (2005). doi: 10.1016/j.aca.2004.12.024
9.
VapnikV, “The support vector method of function estimation”, in Nonlinear Modeling: Advanced Black-box Techniques, Ed by SuykensJ. and VandewalleJ.Springer-Verlag, New York, USA, p. 55 (1998).
10.
SuykensJ.GestelT.De BrabanterJ.De MoorB. and VandewalleJ., Least-Squares Support Vector Machines.World Scientific, Singapore (2002).
11.
CzekajT.WuW. and WalczakB., “About kernel latent variable approaches and SVM”, J. Chemometr.19, 341–354 (2005). doi: 10.1002/cem.937
StoyanovS., Optimization methods and algorithms.Tehnika, Sofia, Bulgaria (1990) (in Bulgarian).
14.
CzekajT.WuW. and WalczakB., “About kernel latent variable approaches and SVM”, J. Chemometr.19, 341 (2005). doi: 10.1002/cem.937
15.
WalshK.GolicM. and GreensillC., “Sorting of fruit using near infrared spectroscopy: Application to a range of fruit and vegetables for soluble solids and dry matter content”, J. Near Infrared Spectrosc.12, 141 (2004). doi: 10.1255/jnirs.419
MukarevM., “Algorithm for fast pruning of LS-SVM regression models”, Scientific conference with inter-national participation, Food Science, Engineering and Technologies 2008 October 2008, University of Food Technologies, Vol. LV, issue 2, Plovdiv, Bulgaria. Collection of Scientific Works of UFT-Plovdiv, Plovidiv, Bulgaria, pp. 123 (2008) (in Bulgarian).
20.
MukarevM., “Pruning of least squares support vector machines regression models”, Proceedings of the Scientific Conference with International Participation: Food Science, Engineering and Technologies2007, Vol. 54, roll 3, 19–20 October 2007, University of Food Technologies, Plovdiv, Bulgaria. Collection of Scientific Works of UFT-Plovdiv, Plovdiv, Bulgaria, pp. 19–24, (2007) (in Bulgarian).
21.
FearnT., “Bias”, NIR news9, 4 (1998).
22.
FearnT., “Comparing standard deviations”, NIR news7, 5–6 (1996).
