Headspace Solid-Phase Microextraction-Gas Chromatography-Mass Spectrometry Quantification of the Volatile Profile of More than 1200 Virgin Olive Oils for Supporting the Panel Test in Their Classification: Comparison of Different Chemometric Approaches

A reliable and robust tool for supporting the panel test in virgin olive oil classification is still required. We propose four chemometric approaches based on t test, principal component analysis (PCA) and linear discriminant analysis (LDA), applied for combining sensorial data, and chemical measurements. The former was from the panel test, and the latter was from headspace solid-phase microextraction-gas chromatography-mass spectrometry quantitation of 73 volatile organic compounds (VOCs) of 1223 typical commercial virgin olive oils, with most of them recognized as difficult to classify with accuracy by the panel test. The approaches were developed and validated, and the best results, with 83.5% correct classification, were using the PCA-LDA approach. Among the other methods, developed for proposing simplified procedures based on a smaller number of VOCs, the best method gave 80.1% correct classification only using 10 VOCs. All of the approaches suggested that octane, heptanal, pent-1-en-3-ol, Z-3-hexenal, nonanal, and 4-ethylphenol should be considered as a basis of volatiles for classification of olive oil samples.