A Methodology Based on FT-IR Data Combined with Random Forest Model to Generate Spectralprints for the Characterization of High-Quality Vinegars

Abstract

Sherry wine vinegar is a Spanish gourmet product under Protected Designation of Origin (PDO). Before a vinegar can be labeled as Sherry vinegar, the product must meet certain requirements as established by its PDO, which, in this case, means that it has been produced following the traditional solera and criadera ageing system. The quality of the vinegar is determined by many factors such as the raw material, the acetification process or the aging system. For this reason, mainly producers, but also consumers, would benefit from the employment of effective analytical tools that allow precisely determining the origin and quality of vinegar. In the present study, a total of 48 Sherry vinegar samples manufactured from three different starting wines (Palomino Fino, Moscatel, and Pedro Ximénez wine) were analyzed by Fourier-transform infrared (FT-IR) spectroscopy. The spectroscopic data were combined with unsupervised exploratory techniques such as hierarchical cluster analysis (HCA) and principal component analysis (PCA), as well as other nonparametric supervised techniques, namely, support vector machine (SVM) and random forest (RF), for the characterization of the samples. The HCA and PCA results present a clear grouping trend of the vinegar samples according to their raw materials. SVM in combination with leave-one-out cross-validation (LOOCV) successfully classified 100% of the samples, according to the type of wine used for their production. The RF method allowed selecting the most important variables to develop the characteristic fingerprint (“spectralprint”) of the vinegar samples according to their starting wine. Furthermore, the RF model reached 100% accuracy for both LOOCV and out-of-bag (OOB) sets.

Detection of Adulterations in Fruit Juices Using Machine Learning Methods over FT-IR Spectroscopic Data

“…With the aim of handling and extracting useful information from these data matrices, the application of data analysis methods (including machine learning methods) is crucial. Thus, both the transformation of data into interpretable information and the fitting of predictive models with interactive applications to automate quality control processes are possible by employing languages of coding, so these global profiling methods are becoming increasingly popular [9] and have been used in various fields such as forensic chemistry [29,30], agri-food [31,32], pharmacological industry [33,34], among others. It is worth it since juices are complex matrices and the differences between spectra are sometimes very substantial, thus identifying a small number of markers can be difficult and unsuccessful.…”

Rapid Detection and Quantification of Adulterants in Fruit Juices Using Machine Learning Tools and Spectroscopy Data

“…Both must be chosen previously to control the balance between bias and variance of the model. For this purpose, an exponential growth search method was used, as described in previous studies [ 41 , 50 , 51 ]. This consists of taking values in the range of −10 to 10 for log 2 C and log 2 γ, and the best combination of the hyperparameters is the one that achieves the highest accuracy value for a 5-fold cross-validation on the training set itself.…”

Enhanced Differentiation of Wild and Feeding Civet Coffee Using Near-Infrared Spectroscopy with Various Sample Pretreatments and Chemometric Approaches

“…The false positive rate (FPR) is the percentage of incorrectly labeled samples, such as wild civet coffee classified as feeding civet coffee or vice versa. The ideal model will strike a balance between high TPR and low FPR [21,25]…”

“…However, using non-parametric classification techniques, specific differences can be found in highly similar samples [19,20]. SVM is used to discriminate different types of coffee based on chemical or spectral features, ensuring the identification and quality control, while RF can handle different data sources and capture non-linear relationships to provide feature importance insights [18,[21][22][23][24]. Hence, it is possible to use NIRS combined with chemometric techniques to differentiate between wild and feeding civet coffee.…”