Skip to main content

Advertisement

Table 5 Classification accuracy results for NNRW classifier under different number of hidden neurons on eye event-related EEG data

From: Feature selection using angle modulated simulated Kalman filter for peak classification of EEG signals

Peak model Result No. of hidden neurons
100 200 300 400 500 600 700 800 900 1000 1100 1200
Dumpala Train 5.15 30.1 43.61 53.39 60.26 66.51 71.27 75.55 78.63 80.86 82.96 84.54
  Test 1.09 15.77 24.83 31.75 38.09 42.12 45.31 48.17 49.37 51.46 52.9 53.87
Acir Train 37.69 48.95 53.37 56.87 59.82 63.27 66.41 70.06 73.69 76.3 79.38 81.73
  Test 34.46 44.05 45.11 46.67 47.74 48.55 49.3 50.2 51.86 52.16 51.67 52.91
Liu Train 35.61 48.54 54.83 60.38 65.41 69.09 71.94 73.99 75.52 77.18 78.62 80.16
  Test 29.18 38.76 41.4 42.97 45.25 46.34 48.07 47.94 48.85 48.19 48.57 48.91
Dingle Train 0 6.19 19 31.13 41.89 49.91 57.07 61.96 68.14 71.39 75.12 77.22
  Test 0 1.55 6.48 15.97 21.97 25.81 32.34 34.78 38.31 40.13 43.65 45.26