Linear and non-linear modeling to identify vinegars in blends through spectroscopic data

摘要: Abstract The identification of vinegars produced from six different raw materials (red wine, white cider, apple, molasses, and rice) in blends has been accomplished through their UV–vis spectra mathematical models: partial least squares discriminant analysis (PLS-DA) artificial neural networks (ANNs). registered were mathematically treated following a linear approach non-linear one (ANN) based on multilayer perceptron models with training functions. average correct classification rate series comparable internal validations was around 55% 90%, for the PLS-DA ANN respectively, which heavily favors approach. Therefore, an accurate chemometric tool ability to detect specific mixtures inexpensive straightforward fashion designed optimized.