Walnut oil (WNO) is a high value vegetable oil with high nutritional value. This study developed a method based on excitation–emission matrix fluorescence (EEMF) spectroscopy coupled with chemometrics and ensemble learning to authenticate WNO, and semi-quantify the adulteration with other vegetable oils. A total of 711 vegetable oil samples were analysed by EEMF of which 426 samples were pure WNO. The classification models were established by chemometrics and ensemble methods. The ensemble methods achieved better classification performance, and PLS (partial least squares) regression was applied to semi-quantify the adulteration levels of walnut oil, especially with rapeseed oil (RSO).
Read the abstract here
