Premium
Application of principal‐component analysis on near‐infrared spectroscopic data of vegetable oils for their classification
Author(s) -
Sato Tetsuo
Publication year - 1994
Publication title -
journal of the american oil chemists' society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.512
H-Index - 117
eISSN - 1558-9331
pISSN - 0003-021X
DOI - 10.1007/bf02638055
Subject(s) - rapeseed , principal component analysis , vegetable oil , cottonseed , food science , fatty acid , chemistry , near infrared spectroscopy , composition (language) , cottonseed oil , coconut oil , edible oil , bran , mathematics , biology , organic chemistry , statistics , raw material , linguistics , philosophy , neuroscience
In the near‐infrared (NIR) spectra of oil, information about fatty acid composition is concentrated in the range of 1600–2200 nm. Principal‐component analysis (PCA) was applied on the standardized full NIR spectral data of this region for vegetable oils to totally capture the NIR spectral pattern. Nine varieties of vegetable oils (soybean, corn, cottonseed, olive, rice bran, peanut, rapeseed, sesame and coconut oil) could be successfully classified from their PCA scores. Examining the contribution of wavelengths to PCA scores showed that wavelengths with a high loading weight were assigned to characteristic absorption regions that correspond to specific fatty acid moieties. This classification is related to the fatty acid composition of an oil, and it can be carried out rapidly and easily after eigenvectors were obtained.