Classification of Motor Imagery EEG Signals with Support Vector Machines and Particle Swarm Optimization

Table 2

The classification results of PSO-SVM and traditional methods for 2005 Data Iva.

Accuracy (%)	⁢aa (100)		⁢al (200)		⁢av (80)		⁢aw (56)
	Max	Average	Max	Average	Max	Average	Max	Average
Decision tree	82.5	79.8	74.6	71.5	74.2	71.2	75.4	72.8
BP	71.4	70.2	85.8	84.2	81.6	79.4	72.2	70.6
KNN	94.6	92.5	87.6	85.8	82.4	80.1	89.2	87.8
LDA	96.2	95.4	92.7	90.2	82.8	80.2	91.6	90.2
SVM	97.7	96.4	92.5	89.4	81.5	77.6	90.5	88.9
PSO-SVM	98.1	97.0	93.9	91.7	82.0	80.5	92.3	91.6