APPLICATION OF ATR-FTIR AND MULTIVARIATE STATISTICS ANALYSIS (MSA) ON CLASSIFICATION OF PEPPER PRODUCTS IN VIET NAM

Abstract

Chemometrics, particularly Multivariate Statistics Analysis techniques including Principle Components
Analysis (PCA) and Linear Discriminate Analysis (LDA) were employed to process spectral data, in
order to classify Pepper products in Vietnamese commercial market. 14 peppercorns (Piper nigrum)
samples (black and white), corresponding to 07 different trademarks were collected in supermarkets,
local markets and food suppliers within Hanoi and then analyzed directly by ATR-FTIR instrument.
Spectral information was simplified by peak selection, before being processed by MSA techniques. In
comparison with other conventional spectroscopy methodologies, ATR-FTIR is considered a rapid and
convenient method due to its non-wasting time of sample preparation. Results obtained have confirmed
the success of trademark identification and classification of pepper products. An obvious separation
between trademark groups was presented on LDA observation although there was still certain
overlapping occurred on the graph; while with PCA, classification between the two sorts of peppercorn
is still an issue that needs further study.