The potential of coupling information using three analytical techniques for identifying the geographical origin of Liguria extra virgin olive oil

Abstract

Abstract The combined use of data obtained from different analytical instruments is a complex problem. In this paper the potential of coupling three analytical techniques for building a class model for extra virgin (e.v.) olive oil from Liguria, was investigated. A sampling design for the Ligurian e.v. olive oil was developed by the selection of a representative subset from all possible e.v. olive oil samples of the Liguria region. Thus, in order to choose this subset with uniform distribution on the production area and representative of the production density, two algorithms for sampling have been used: Kennard–Stone and Potential Function. The samples were analysed by head-space mass spectrometry (electronic nose), UV–visible and NIR spectroscopy. In particular, the exceptional possibility provided by Chemometrics to effectively extract and combine (fusion) the information from these multivariate and non-specific data and to build a class model for Liguria e.v. olive oil was studied.