Research Article
Multiple Kernel Dimensionality Reduction via Ratio-Trace and Marginal Fisher Analysis
Table 3
The learned weights and classification accuracies of KFDA with the single kernel on Ionosphere.
| | Weights learned by MKL-TR | Weights learned by MKL-MFA | KFDA with the single kernel |
| 0.10 | 0.02 | 0.01 | 64.53 | 0.22 | 0.02 | 0.01 | 65.75 | 0.46 | 0.06 | 0.03 | 90.82 | 1.00 | 0.36 | 0.40 | 92.24 | 2.15 | 0.37 | 0.42 | 93.07 | 4.46 | 0.13 | 0.11 | 91.98 | 10.00 | 0.03 | 0.02 | 90.17 | 21.54 | 0.02 | 0.01 | 88.67 | 46.42 | 0.001 | 0.001 | 87.73 | 100.00 | 0.01 | 0.002 | 85.26 |
|
|