PLS-DA classification of moroccan olive varieties by using direct ATR-FTIR analysis on fresh olives

Food traceability implies the control of the entire chain of food production and marketing, allowing the food to be traced through every step of its production back to its origin. Traceability is particularly relevant to assess origin and content of the olive oil but also to protect and prevent frauds at different stages in oil production. The quality of virgin olive oil is principally function of four parameters: variety, geographic origin, fruit quality and extraction process. The purpose of this work was to characterize and classify five Moroccan varieties of olives by combining Fourier transform infrared spectroscopy and chemometrics [1-2]. Attenuated total reflectance was adopted as sampling technique in conjunction with infrared spectroscopy, allowing the solid state analysis of olive endocarps without any preliminary treatment. Partial least squares-discriminant analysis was performed to assess the classification capacity of the FTIR data among the five cultivars. The spectral data were subjected to a preliminary derivative elaboration based on the Norris gap algorithm to reduce the noise and extract larger analytical information. The classification model was optimized by applying the Martens' uncertainty test which provided to select the wavelength zones giving the most useful information. Application of PLS-DA allowed to classify all the samples into five variety groups with a correct classification of 84.2% when the model was applied on the prediction set. [1] M. De Luca, W. Terouzi, G. Ioele et al. Derivative FTIR spectroscopy for cluster analysis and classification of morocco olive oils. Food Chemistry 124 (2011) 1113-1118. [2] W. Terouzi, M. De Luca, A. Bolli et al. A discriminant method for classification of Moroccan olive varieties by using FT-IR analysis of the mesocarp section.Vibrational Spectroscopy 56 (2011) 123-128