LAUSR

Solid-phase microextraction - mass spectrometry (SPME-MS) and fast gas chromatography based electronic nose (GC-E-Nose) were used and compared for their suitability to distinguish Baijiu of various aroma types and geographical origin. Baijiu is a traditional Chinese distilled spirit produced with complex consortia of microorganisms, which results in very complex aroma compounds profiles. A total of 65 Baijiu samples representing 6 aromas were investigated. Strong aroma types from 3 regions were examined for their origin. Data acquired on two analytical systems were processed using uniform statistical approach. Data were pre-processed for multi-classification (OPLS-DA) models as well as for binary classification (PLS-DA) ones. Aroma and regional classification performed using OPLS-DA indicated that the approach based on SPME-MS had better fitness and prediction ability compared with GC-E-Nose. The total correct classification rate for SPME-MS was 94.44% for aroma and 100% for region, whereas for GC-E-Nose these values were 91.53% and 93.94% respectively.