Research Article
Near Infrared Spectroscopy Based on Supervised Pattern Recognition Methods for Rapid Identification of Adulterated Edible Gelatin
Table 2
Predicting results of the LDA model in the training set and validation set.
| Category | Training set | Validation set | Recognition rate | Rejection rate | Recognition rate | Rejection rate |
| Pure edible gelatin | 100% (11/11) | 100% (28/28) | 90.91% (10/11) | 100% (28/28) | 4 : 1 adulterated gelatin | 100% (7/7) | 100% (32/32) | 100% (7/7) | 96.88% (31/32) | 3 : 2 adulterated gelatin | 100% (5/5) | 100% (34/34) | 100% (5/5) | 100% (34/34) | 2 : 3 adulterated gelatin | 100% (7/7) | 100% (32/32) | 100% (7/7) | 100% (32/32) | 1 : 4 adulterated gelatin | 100% (6/6) | 100% (33/33) | 100% (6/6) | 100% (33/33) | Pure industrial gelatin | 100% (3/3) | 100% (36/36) | 100% (3/3) | 100% (36/36) |
|
|