A rapid method to discriminate season of production and feeding regimen of butters based on infrared spectroscopy and artificial neural networks

Herein, the potential of Fourier transform infrared spectroscopy (FTIR) spectroscopy coupled with principal component analysis and artificial neural networks to discriminate butters obtained from milk creams collected in different seasons (spring, summer or winter) and produced from two feeding regimens (traditional or unifeed) in the Parmigiano Reggiano cheese area was investigated. The predictive ability of neural networks to predict the season of production was 100%. The favorable results obtained by artificial neural network (ANN) analysis were in agreement with those reported by principal component analysis (PCA) analysis. The ability of ANNs to predict the feeding regimens was 90.0%, 75.0% and 75.0%, respectively, for samples collected in spring, summer and winter. These results also confirm that the method is highly suitable for its intended purpose. Highlights ► 405 FT-IR spectra of butter were analyzed by PCA and ANN. ► ANN discriminate the season of production and dietary regimen of butters. ► The ability of ANN to predict the season of production of milk cream was 100%.