Research Article
Feature Selection and Classifier Parameters Estimation for EEG Signals Peak Detection Using Particle Swarm Optimization
Table 7
Training and testing performance of peak detection for each peak model (without feature selection).
| Peak model | Training (%) | Testing (%) | Average | Max | Min | STDEV | Average | Max | Min | STDEV |
| Dumpala et al. (1982) [8] | 84.01 | 89.15 | 80.58 | 4.43 | 81.22 | 91.83 | 74.15 | 9.13 | Acir et al. (2005) [7, 11, 26] | 74.4 | 80.59 | 67.08 | 3.71 | 68.59worst | 77.43 | 54.77 | 6.97 | Liu et al. (2002) [10] | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | Dingle et al. (1993) [9] | 90.98 | 94.76 | 83.66 | 5.1 | | 94.75 | 77.44 | 7.98 |
|
|