Premium
Identity Confirmation of a Branded, Fermented Cereal Product by UV Spectroscopy: A Feasibility Study Involving a Trappist Beer
Author(s) -
Weeranantanaphan Jittima,
Downey Gerard
Publication year - 2010
Publication title -
journal of the institute of brewing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.523
H-Index - 51
eISSN - 2050-0416
pISSN - 0046-9750
DOI - 10.1002/j.2050-0416.2010.tb00398.x
Subject(s) - brand identity , chemistry , partial least squares regression , statistics , mathematics , business , marketing
Brand protection is important for a food processor; trust in brand identity is essential for consumer confidence. In this study, the combination of UV spectroscopy and multivariate mathematics was investigated to confirm the identity of a processed cereal product—a Trappist beer. Samples (52) of Rochefort 8, other Rochefort and non‐Rochefort beers, diluted 1:100 with distilled water, were analysed by UV spectroscopy (220–400 nm). Partial least squares discriminant analysis (PLS‐DA) and soft independent modelling of class analogy (SIMCA) multivariate methods were applied separately to confirm the identity of Rochefort 8 beer. Spectral data were analysed in both raw and standard normal variate (SNV) pre‐treated forms. Using PLS‐DA, a two‐stage modelling procedure was applied involving initial classification as either Rochefort or non‐Rochefort followed by classification within the Rochefort class as Rochefort 8 or non‐Rochefort 8. Correct classification rates for these two steps were (a) 100 and 94.4% and (b) 90.9 and 100% respectively. Applying the 3 principal component SIMCA model to the 8 test Rochefort and all of the non‐Rochefort (n = 36) samples, 6 of the 8 Rochefort beers were correctly classified (75%; p = 0.05) but only two of the 36 non‐Rochefort beers were wrongly classified as Rochefort (correct classification rate of 94.4%).