Fourier transform infrared (FTIR) and fluorescence spectroscopy combined with soft independent modeling
of class analogies (SIMCA) and partial least square (PLS) were used to detect the authenticity of walnut
oil and adulteration amount of soybean oil in walnut oil. A SIMCA model of FTIR spectra could
differentiate walnut oil and other oils into separate categories; the classification limit of soybean oil in
walnut oil was 10%. Fluorescence spectroscopy could differentiate oil composition by the peak position
and intensity of emission spectrum without multivariate analysis. The classification limit of soybean
oil adulterated in walnut oil by fluorescence spectroscopy was below 5%. The deviation of the prediction
model for fluorescence spectra was lower than that for FTIR spectra. Fluorescence spectroscopy was more
applicable than FTIR in the adulteration detection of walnut oil, both from the determination limit and
prediction deviation