Research Article

Investigating Feature Ranking Methods for Sub-Band and Relative Power Features in Motor Imagery Task Classification

Table 1

Classification accuracy of different classifiers.

SubjectsClassification accuracy (in %)
TreeFine KNNWeighted KNNQuadratic SVMRandom forest

S0189.192.994.365.389.4
S0287.391.992.870.591.1
S0389.592.994.27191.34
S0488.290.693.667.390.625
S0585.591.893.166.186.77
S0687.790.293.158.490.86
S0788.789.191.66987.01
S0886.791.392.964.489.66
S099292.795.273.792.78
S1087.792.994.367.591.34
AVG88.2491.6393.5167.3290.0885

Bold letters show the maximum classification accuracy of the classifier.