Research Article
Classification of Motor Imagery EEG Signals with Support Vector Machines and Particle Swarm Optimization
Table 3
The classification results of PSO-SVM and traditional methods for 2008 Dataset 1.
| Accuracy (%) | a | b | f | e | Max | Average | Max | Average | Max | Average | Max | Average |
| Decision tree | 84.5 | 81.4 | 74.6 | 72.5 | 86.2 | 84.5 | 84.6 | 78.2 | BP | 85.6 | 82.2 | 74.8 | 73.2 | 87.6 | 85.4 | 83.4 | 81.5 | KNN | 90.6 | 86.5 | 78.6 | 76.8 | 91.4 | 87.1 | 84.2 | 82.8 | LDA | 89.4 | 86.2 | 80.5 | 76.6 | 90.8 | 86.0 | 86.4 | 82.2 | SVM | 91.7 | 86.8 | 78.0 | 76.3 | 95.0 | 86.1 | 90.5 | 78.3 | PSO-SVM | 91.3 | 88.1 | 83.5 | 80.1 | 95.2 | 89.7 | 92.0 | 83.1 |
|
|