Research Article
Near Infrared Spectroscopy Based on Supervised Pattern Recognition Methods for Rapid Identification of Adulterated Edible Gelatin
Table 5
Optimization parameters and predicting results of SVM models based on three parameter optimization algorithms.
| Parameter optimized methods | Best C | Best γ | CV accuracy | Recognition rate | Training set | Validation set |
| GS-SVM | 0.3299 | 5.2780 | 100% | 100% (75/75) | 100% (39/39) | GA-SVM | 1.1135 | 11.4377 | 100% | 100% (75/75) | 100% (39/39) | PSO-SVM | 0.5283 | 15.7922 | 100% | 100% (75/75) | 100% (39/39) |
|
|