Comparison of Multivariate Calibration and Discriminant Analysis in Evaluating NIR Spectroscopy for Determination of Meat Tenderness
Authors:
- 1 March 1997
- journal article
- Published by SAGE Publications in Applied Spectroscopy
- Vol. 51 (3) , 350-357
Abstract
Often the primary goal of analytical measurement tasks is not to find good estimates of continuous reference values but rather to determine whether a sample belongs to one of a number of categories or subgroups. In this paper the potential of different statistical techniques in the classification of raw beef samples in tenderness subgroups was studied. The reference values were based on sensory analysis of beef tenderness of 90 samples from bovine M. longissimus dorsi muscles. The sample set was divided into three categories—very tough, intermediate, and very tender—according to degree of tenderness. A training set of samples was used to find the relationship between category and near-infrared (NIR) spectroscopic measurements. The study indicates that classical discriminant analysis has advantages in comparison to multivariate calibration methods [i.e., principal component regression (PCR)], in this application. One reason for this observation seems to be that PCR underestimates high measurement values and overestimates low values. In this way most samples are assigned to the intermediate group of samples, causing a small number of erroneous classifications for the intermediate subgroup, but a large number of errors for the two extreme groups. With the use of PCR the number of correct classifications in the extreme subgroups was as low as 23%, while the use of discriminate analysis increased this number to almost 60%. The number of classifications in correct or neighbor subgroup for the two extreme subgroups was equal to 97%. A “bias-correction” was also attempted for PCR, and this gave results comparable to the best results obtained by discriminant analysis methods. Test sets used NIR analysis of fresh, raw beef samples with different processing. While this spectroscopic approach had previously been shown to be useful with frozen products, it appears unsuitable at this time for fresh beef. However, its marginal analytical utility proved useful in evaluating the two classification approaches employed in this study.Keywords
This publication has 8 references indexed in Scilit:
- Prediction of sensory characteristics of beef by near-infrared spectroscopyMeat Science, 1994
- Determination of the Sodium Chloride Content of Sausages by near Infrared SpectroscopyJournal of Near Infrared Spectroscopy, 1993
- Stepwise canonical discriminant analysis of continuous digitalized signals: Application to chromatograms of wheat proteinsJournal of Chemometrics, 1990
- Prediction of sensory quality by near infrared reflectance analysis of frozen and freeze dried green peas (Pisum sativum)Journal of the Science of Food and Agriculture, 1990
- Principal components of a fuzzy clusteringTrAC Trends in Analytical Chemistry, 1989
- Classification: Oldtimers and newcomersJournal of Chemometrics, 1989
- Prediction of gasoline octane numbers from near-infrared spectral features in the range 660-1215 nmAnalytical Chemistry, 1989
- Detection and Characterization of Cluster Substructure I. Linear Structure: Fuzzy c-LinesSIAM Journal on Applied Mathematics, 1981