Characterization of commercial chili sauce varieties according to their chemical and physical properties using chemometric methods

Different varieties of chili sauces that are commercially available in Malaysia were randomly collected and their physical and chemical parameters were determined. Initially, pattern recognition techniques were applied in order to categorize the chili sauce varieties. With the aid of principal component analysis and cluster analysis, it is possible to visualize the clustering tendencies of the different varieties of chili sauces where four major clusters, namely chili sauces, general chili ketchups, hot/garlic added ketchups, and Thai ketchups have been successfully partitioned. Color and taste parameters have been found to be the most discriminating variables among them. With the K-nearest neighbor technique, a good prediction of chili sauces' categories could be achieved. With appropriate customer preference survey, it would be possible to map these characteristics to the preferences. This would certainly help producers of chili sauces in upgrading the quality and taste of their products. © 2009 Springer Science+Business Media, LLC