abstract
- Legal regulations are set for protecting claims regarding olive oil geographical denomination. When meteorological or agroecological factors similarly affect different regions, the origin identification is a challenging task. This study demonstrated the use of a potentiometric electronic tongue coupled with linear discriminant analysis to discriminate the geographical origin of monovarietal Tunisian olive oil produced from local cv Chemlali (Kairouan, Sidi Bouzid or Sfax regions) and cv Sahli (Kairouan, Mahdia or Sousse regions). The potentiometric fingerprints of 12 or eight lipid sensors (for Chemlali and Sahli, respectively), selected using a simulated annealing meta-heuristic algorithm, allowed the correct prediction (repeated K-fold cross-validation) of the geographic production region with sensitivities of 92 ± 7% (Chemlali) and 97 ± 8% (Sahli). It was also confirmed the electronic tongue capability to classify Tunisian olive oil according to olive cultivar or quality grade. The results indicated the possible use of potentiometric fingerprints as a promising innovative strategy for olive oil analysis allowing assessing geographical origin, olive cultivar and quality grade, which are key factors determining olive oil price and consumers’ preference.