Table 7.
Modelling performance of PLSDA and SVM classifiers built from one concentration and applied to other concentrations (deposited on STS) using 3500–2600 cm−1.
| Applied to | 10 OD | 1 OD | 0.1 OD | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Built from | LVs | OA | MCC | Sen | Spe | OA | MCC | Sen | Spe | OA | MCC | Sen | Spe | ||
| PLSDA | 10 OD | 10 | 100% | 1.00 | 1.00 | 1.00 | 91% | 0.83 | 0.98 | 0.84 | 75% | 0.50 | 0.98 | 0.40 | |
| 1 OD | 6 | 87% | 0.77 | 1.00 | 0.75 | 95% | 0.91 | 0.95 | 0.95 | 82% | 0.62 | 0.94 | 0.62 | ||
| 0.1 OD | 7 | 73% | 0.46 | 0.72 | 0.73 | 89% | 0.79 | 0.85 | 0.93 | 93% | 0.85 | 0.95 | 0.90 | ||
| SVM | 10 OD | - | 100% | 1.00 | 1.00 | 1.00 | 89% | 0.81 | 0.99 | 0.81 | 75% | 0.52 | 1.00 | 0.38 | |
| 1 OD | - | 86% | 0.75 | 0.99 | 0.73 | 98% | 0.96 | 0.98 | 0.98 | 83% | 0.65 | 0.95 | 0.66 | ||
| 0.1 OD | - | 69% | 0.40 | 0.51 | 0.87 | 87% | 0.75 | 0.79 | 0.94 | 95% | 0.90 | 0.98 | 0.92 | ||
OA: overall accuracy; MCC: Matthews correlation coefficient; Sen: sensitivity; Spe: specificity.