Research Article
The Classification of Valid and Invalid Beats of Three-Dimensional Nystagmus Eye Movement Signals Using Machine Learning Methods
Table 3
The parameter values for the kernel functions that produced the best classification results with support vector machines.
| ā | Linear kernel: box constraint | Quadratic kernel: box constraint |
Radial basis function: box constraint and |
| Manual selection | 0.5 | 9.5 | 2.5 | 0.5 | Manual selection with cleaning | 3.5 | 0.5 | 3.0 | 10.0 | Automatic selection | 7.5 | 3.5 | 8.5 | 0.5 | Automatic selection with cleaning | 2.5 | 8.0 | 6.5 | 10.0 | Both manual and automatic selections | 1.5 | 1.0 | 0.5 | 0.5 | Both selections with cleaning | 1.0 | 0.5 | 7.5 | 10.0 |
|
|