Classification of chilli sauces: multivariate pattern recognition using selected gcms retention time peaks of chilli sauce samples
Document Type
Article
Publication Date
1-1-2008
Abstract
As a preliminary work on the possibility of separating classes of chili sauces based on taste or customer preferences, organic compounds from different kinds of chili sauces of various brands were separated and analyzed by gas chromatography/mass spectrometry (GC/MS). It was found that these organic compounds do form a basis for separation of different types of sauces. The similarity and dissimilarity of chromatograms due to the organic composition of the chili sauces were explored by multivariate pattern recognition techniques based on cluster analysis (CA) and principal component analysis (PCA). Both CA and PCA results exhibit four linearly separable classes, namely general sauces, hot sauces, sauces with benzoic acid and sauces with garlic. It was concluded that by using chosen retention peaks in the chromatograms of various sauce samples as multivariate features, CA and PCA can be successfully used to reveal the natural clusters existing in chili sauces according to their organic composition.
Keywords
Chemometrics, Chili sauce, Principal component analysis, PCA, Cluster analysis
Divisions
CHEMISTRY
Publication Title
Malaysian Journal of Analytical Sciences
Volume
12
Issue
1
Additional Information
Department of Chemistry, Faculty of Science Building, University of Malaya, 50603 Kuala Lumpur, MALAYSIA