Research Article
Improving Localized Multiple Kernel Learning via Radius-Margin Bound
Table 1
Classification accuracy (mean ± standard derivation) on the selected datasets.
| Dataset | SimpleMKL | LMKL | ILMKL |
| Ionosphere | 92.07 ± 2.02 | 91.64 ± 2.77 | 93.89 ± 1.21 | Heart | 84.01 ± 1.38 | 83.92 ± 1.65 | 85.91 ± 2.84 | Wdbc | 96.19 ± 0.44 | 95.54 ± 2.35 | 96.97 ± 1.96 | Wpbc | 76.59 ± 1.02 | 77.40 ± 2.21 | 78.42 ± 1.14 | Liver | 64.45 ± 3.64 | 62.13 ± 2.93 | 64.87 ± 3.46 | Musk1 | 89.79 ± 2.02 | 86.44 ± 3.37 | 90.27 ± 4.68 | Sonar | 83.42 ± 3.30 | 79.95 ± 3.15 | 82.98 ± 5.02 | Fourclass | 99.95 ± 0.06 | 99.93 ± 0.14 | 99.97 ± 0.09 | Coloncancer | 74.98 ± 8.18 | 74.83 ± 3.93 | 76.11 ± 1.16 | Germannumber | 72.74 ± 1.55 | 73.32 ± 1.78 | 75.24 ± 1.15 | Pima | 74.37 ± 2.05 | 75.23 ± 3.12 | 77.01 ± 1.52 |
|
|