作者: P. Oliveri , V. Di Egidio , T. Woodcock , G. Downey
DOI: 10.1016/J.FOODCHEM.2010.10.047
关键词:
摘要: Following the introduction of legal identifiers geographic origin within Europe, methods for confirming any such claims are required. Spectroscopic techniques provide a method rapid and non-destructive data collection variety chemometric approaches have been deployed their interrogation. In this present study, class-modelling (SIMCA, UNEQ POTFUN) after compression by principal component analysis development class-models set olive oils honeys. The number components, confidence level spectral pre-treatments (1st 2nd derivative, standard normal variate) were varied, strategy variable selection was tried. Models evaluated on separate validation sample set. outcomes reported criteria most appropriate models given application discussed.