Research Article
The Impact of Simulated Spectral Noise on Random Forest and Oblique Random Forest Classification Performance
Table 3
oRF-SVM classification accuracies for three simulated noise levels for three weeks of infestation.
| Noise (%) | Week one | Week two | Week three | 0 | 10 | 20 | 30 | 0 | 10 | 20 | 30 | 0 | 10 | 20 | 30 |
| Train accuracy | 96.19 | 97.80 | 94.71 | 94.71 | 96.42 | 97.08 | 95.60 | 96.31 | 93.35 | 94.11 | 94.11 | 91.13 | Test accuracy | 94.64 | 91.07 | 89.28 | 87.50 | 94.64 | 96.42 | 94.64 | 94.64 | 100.00 | 100.00 | 100.00 | 100.00 | Train kappa | 92.30 | 95.57 | 89.23 | 89.33 | 92.85 | 94.14 | 91.26 | 92.69 | 86.66 | 88.23 | 88.23 | 82.28 | Test kappa | 89.28 | 82.14 | 78.57 | 75.00 | 89.28 | 92.85 | 89.28 | 89.28 | 100.00 | 100.00 | 100.00 | 100.00 |
|
|