Jump to main content
Jump to site search

Issue 9, 2012
Previous Article Next Article

Quantitative analysis of adulteration of extra virgin olive oil using Raman spectroscopy improved by Bayesian framework least squares support vector machines

Author affiliations

Abstract

The adulteration of extra virgin olive oil (EVOO) is a big problem in food safety. The present paper uses Raman spectra to characterize different kinds of vegetable oils in the region 800–1800 cm−1. Bayesian framework is applied to find the best parameters for the least squares support vector machines (LS-SVM), and an adulteration prediction model is established by using the optimal parameters and the Raman spectral data of EVOO for the training of LS-SVM without any classification process. The results show that the root mean square error of prediction (RMSEP) and the coefficient of determination (R2) of the algorithm based on Bayesian framework LS-SVM (Bay-LS-SVM) are 0.0509 and 0.9976, respectively. Compared with the commonly used chemometric tool, partial least squares regression (PLS), the proposed algorithm shows higher accuracy and computational efficiency. The method based on Bay-LS-SVM and Raman spectroscopy is also easy to operate, non-destructive and ‘lipid sensitive’, and it is considered to be suitable for online detection of adulterated olive oil.

Graphical abstract: Quantitative analysis of adulteration of extra virgin olive oil using Raman spectroscopy improved by Bayesian framework least squares support vector machines

Back to tab navigation

Supplementary files

Publication details

The article was received on 27 Apr 2012, accepted on 02 Jun 2012 and first published on 09 Jul 2012


Article type: Paper
DOI: 10.1039/C2AY25431J
Citation: Anal. Methods, 2012,4, 2772-2777
  •   Request permissions

    Quantitative analysis of adulteration of extra virgin olive oil using Raman spectroscopy improved by Bayesian framework least squares support vector machines

    W. Dong, Y. Zhang, B. Zhang and X. Wang, Anal. Methods, 2012, 4, 2772
    DOI: 10.1039/C2AY25431J

Search articles by author

Spotlight

Advertisements