Premium
Discrimination and Cluster Analysis of Soy Sauce GLC Profiles
Author(s) -
AISHIMA TETSUO
Publication year - 1982
Publication title -
journal of food science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.772
H-Index - 150
eISSN - 1750-3841
pISSN - 0022-1147
DOI - 10.1111/j.1365-2621.1982.tb04983.x
Subject(s) - aroma , linear discriminant analysis , cluster (spacecraft) , chemistry , chromatography , mathematics , food science , statistics , computer science , programming language
Two hundred chromatograms of volatiles in eight brands of soy sauce were analyzed by discriminant and cluster analysis in order to investigate the difference in gas chromatographic profiles among brands. The sensory tests clearly showed differences in aroma quality among brands. Most of the gas chromatograms were correctly classified by stepwise discriminant analysis into their original brands with a few percent of misclassification. After 48 samples were selected by a random process, cluster analysis was performed. The gas chromatograms in each brand combined to make six or seven clusters on the basis of the selected ten peaks. The connection order of clusters shifted according to the peaks selected.