Research Article
Classification of Electroencephalogram Signal for Developing Brain-Computer Interface Using Bioinspired Machine Learning Approach
Table 1
Performances of subjects of the age group 20–28 using the AR Yule-Walker features with FFNNCSA.
| Subjects | Ten trials’ average training time (sec) | Ten trials’ average testing time (sec) | Average classification accuracy (%) | Sd | Max | Min | Mean |
| S1 | 18.79 | 0.76 | 1.87 | 96.82 | 89.95 | 94.78 | S2 | 18.62 | 0.78 | 1.73 | 96.67 | 89.50 | 93.54 | S3 | 18.30 | 0.72 | 1.38 | 98.34 | 90.72 | 95.42 | S4 | 18.32 | 0.77 | 1.53 | 96.23 | 90.16 | 94.10 | S5 | 18.95 | 0.73 | 1.68 | 96.38 | 89.66 | 94.38 | S6 | 18.68 | 0.75 | 1.57 | 96.74 | 89.82 | 94.66 | S7 | 18.44 | 0.76 | 1.62 | 96.80 | 90.00 | 94.68 | S8 | 18.56 | 0.74 | 1.70 | 95.89 | 89.58 | 94.40 | S9 | 18.78 | 0.72 | 1.71 | 95.92 | 89.76 | 94.84 | S10 | 18.18 | 0.71 | 1.34 | 98.76 | 90.85 | 95.78 |
|
|