Table 2 Comparative performance of different kernels for SVM-based classification of wheat genotypes.
From: Ensemble and optimization algorithm in support vector machines for classification of wheat genotypes
Kernel | Number of Support vectors | Training | Testing | ||
|---|---|---|---|---|---|
Accuracy | Kappa | Accuracy | Kappa | ||
Linear | 67 | 0.934 | 0.899 | 0.932 | 0.896 |
RBF | 190 | 0.938 | 0.905 | 0.932 | 0.897 |
Sigmoid | 143 | 0.815 | 0.714 | 0.864 | 0.790 |
Degree-1 polynomial | 145 | 0.930 | 0.892 | 0.847 | 0.765 |
Degree-2 polynomial | 211 | 0.761 | 0.620 | 0.610 | 0.382 |
Degree-3 polynomial | 195 | 0.918 | 0.871 | 0.780 | 0.648 |
EWA approach | – | 0.951 | 0.924 | 0.949 | 0.922 |
