Table 4.
Spectroscopic Technique | Spectral Region | Parameters for Authentication |
Classification
Method 1 |
Remark | Reference |
Synchronous fluorescence spectroscopy | λex = 250–350 nm and λem = 250–500 nm | Discrimination of white wines according to variety in Tokaj (Slovakia) | PCA, LDA | Correct classification of variety, 100% | [46] |
Total fluorescence spectroscopy | EEM λex =240–800 nm and λem 242–824 nm |
Discrimination of Cabernet Sauvignon wines from Australia and Bordeaux, France | SVMDA XGBDA |
Correct classification of geographical origin using: XGBDA, 100%; and SVMDA, 85% | [11] |
Total fluorescence spectroscopy | EEM λex = 250–500 nm and λem 275–600 nm |
Discrimination of white wine from Romania and France for geographical origin and variety | PARAFAC, SIMCA | Correct classification of: variety, 97%; and geographical origin, 98% | [47] |
Total fluorescence spectroscopy | EEM λex =240–700 nm and λem 242–824 nm |
Discrimination of red wine varieties from different Australian regions for variety and geographical origin | XGBDA | Correct classification of: variety, 100%; and geographical origin, 99.7% | [48] |
1 PCA, principal component analysis; LDA, linear discriminant analysis; SVMDA, support vector machine discriminant analysis; XGBDA, extreme gradient boosting discriminant analysis; PARAFAC, parallel factor analysis; SIMCA, soft independent modelling of class analogy.