Electronic Nose in Combination with Chemometrics for Characterization of Geographical Origin and Agronomic Practices of Table Grape

Nowadays, the protection of food quality attributes (such as geographical origin or method of production) from frauds and adulterations is one of the main concerns of honest producers and aware consumers. In this study, table grape samples were analyzed by using an electronic nose aiming to evaluate the usefulness of sensor data in combination with statistical analysis in discriminating the agronomic practice (conventional vs. organic farming) and the geographical origin of table grape. Principal component analysis (PCA) showed inadequate clustering of samples according to places of production or agronomic practice; thus for classification purpose, a supervised approach was carried out. In particular, linear discriminant analyses (LDA) was used, resulting in mean prediction abilities of 83.6% and 84.6% for the discrimination of farming method and geographical origin, respectively. Considering the results obtained herein, it can be concluded that sensor data combined with chemometrics showed a good potential in discriminating origin as well as method of production of table grapes especially if compared with other analytical techniques both in terms of time and cost of analyses.