Table 3.
Specificity, recall, Precision, F1-score, accuracy and kappa for the classification of SVMs for the external validation (n = 30) of fresh samples using Vis and MSI data and of frozen-thawed samples using MSI data considering five classes from 0% beef-100% offal (0%) to 100% beef-0% offal (100%).
| True Class | ||||||
|---|---|---|---|---|---|---|
| Sensors | 0% | 25% | 50% | 75% | 100% | |
| Vis fresh samples |
Specificity (%) | 100.00 | 70.83 | 100.00 | 100.00 | 100.00 |
| Recall (%) | 100.00 | 100.00 | 0.00 | 83.33 | 100.00 | |
| Precision (%) | 100.00 | 46.15 | 1 NaN | 100.00 | 100.00 | |
| F1-score | 1.00 | 0.63 | 1 NaN | 0.91 | 1.00 | |
| Accuracy (%) | 76.67 | |||||
| Kappa | 0.62 | |||||
| MSI fresh samples |
Specificity (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 |
| Recall (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | |
| Precision (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | |
| F1-score | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | |
| Accuracy (%) | 100.00 | |||||
| Kappa | 1.00 | |||||
| MSI frozen-thawed samples |
Specificity (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 |
| Recall (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | |
| Precision (%) | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | |
| F1-score | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | |
| Accuracy (%) | 100.00 | |||||
| Kappa | 1.00 | |||||
1 NaN: Not a Number (i.e., the outcome of division by 0).