Prediction of olive oil sensory descriptors using instrumental data fusion and partial least squares (PLS) regression Eva Borràs a , Joan Ferré b , Ricard Boqué b , Montserrat Mestres a , Laura Aceña a , Angels Calvo c , Olga Busto a,n a iSens Group, Department of Analytical Chemistry and Organic Chemistry, Universitat Rovira i Virgili, Campus Sescelades, 43007 Tarragona, Spain b Chemometrics, Qualimetrics and Nanosensors Group, Department of Analytical Chemistry and Organic Chemistry, Universitat Rovira i Virgili, Campus Sescelades, 43007 Tarragona, Spain c Official Taste Panel of Virgin Olive Oil in Catalonia, Reus, Spain article info Article history: Received 23 March 2016 Received in revised form 14 April 2016 Accepted 19 April 2016 Available online 20 April 2016 Keywords: Olive oil Electronic panel Mass spectrometry MIR spectroscopy UV–vis spectrophotometry Data fusion Sensory evaluation Partial least squares (PLS) Regression Multivariate analysis abstract Headspace-Mass Spectrometry (HS-MS), Fourier Transform Mid-Infrared spectroscopy (FT-MIR) and UV- Visible spectrophotometry (UV–vis) instrumental responses have been combined to predict virgin olive oil sensory descriptors. 343 olive oil samples analyzed during four consecutive harvests (2010–2014) were used to build multivariate calibration models using partial least squares (PLS) regression. The re- ference values of the sensory attributes were provided by expert assessors from an official taste panel. The instrumental data were modeled individually and also using data fusion approaches. The use of fused data with both low- and mid-level of abstraction improved PLS predictions for all the olive oil de- scriptors. The best PLS models were obtained for two positive attributes (fruity and bitter) and two defective descriptors (fusty and musty), all of them using data fusion of MS and MIR spectral fingerprints. Although good predictions were not obtained for some sensory descriptors, the results are encouraging, specially considering that the legal categorization of virgin olive oils only requires the determination of fruity and defective descriptors. & 2016 Elsevier B.V. All rights reserved. 1. Introduction Olive oil is obtained from the fresh fruit of the olive tree (Olea europaea L.) exclusively by mechanical or other physical means [1]. It is a fundamental ingredient of the Mediterranean diet that is usually consumed in its crude form. The olive oil unique aroma and delicate flavour (mainly conferred by minor compounds), to- gether with its claimed health benefits [2], has contributed to an increase in the global demand for this food commodity. As a consequence, olive oil prices have increased, encouraging some farmers and producers to practice fraudulent activities, such as mixing high quality olive oils with cheaper vegetable oils or with lower quality olive oils. Some international institutions such as the European Union, the International Olive Council and the Codex Alimentarius have adopted a series of regulations to detect pos- sible adulterations and to guarantee olive oil quality and safety [1,3]. These control measures have made olive oil one of the most strictly regulated food products [4], with defined physico-chemical parameters (free acidity, peroxide value, fatty acids and specific ultraviolet (UV) absorptions) as well as present organoleptic characteristics determined by official methods. According to maximum values that cannot be exceeded, these parameters dif- ferentiate three main quality categories that determine the olive oil economic value (Table 1). The two first categories, extra virgin (EVOO) and virgin (VOO) olive oils can be bottled and directly consumed, while lampante olive oil (LOO) must be previously re- fined [5]. The sensory assessment plays a crucial role in the determina- tion of olive oil categories. The only homologated method is per- formed by the taste panel, which employs well-standardized protocols as well as continuously and well-trained panelists. The main task of the taste panel is to evaluate the so-called positive and negative sensory attributes. The positive notes, perceived by con- sumers as “healthy” indicators [8], are mainly attributable to fruity (green or ripe), bitter and pungent sensations. Reminiscent sen- sation of freshly cut grass (grassy), green fruits (green odor), sweet and astringent notes can be also considered as positive char- acteristics of good quality olive oils. On the contrary, winery-vi- negary, fusty, mustiness-humidity, rancid and metallic are the Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/talanta Talanta http://dx.doi.org/10.1016/j.talanta.2016.04.040 0039-9140/& 2016 Elsevier B.V. All rights reserved. n Corresponding author. E-mail address: olga.busto@urv.cat (O. Busto). Talanta 155 (2016) 116–123