Skip to main content

Table 5 Summary of the best results obtained for all the main methodological options considered

From: A knowledge discovery methodology from EEG data for cyclic alternating pattern detection

 

Multi-class problem

Binary problem

mRMR

PCA

mRMR

PCA

#F.

AC (%)

WAC (%)

#P. C.

AC (%)

WAC (%)

#F.

AC (%)

WAC (%)

#P. C.

AC (%)

WAC (%)

DA

30

61

47

12

68

49

43

67

72

15

73

76

k-NN

30

70

46

40

61

47

55

75

80

54

68

75

SVM

40

71

51

30

56

47

40

76

78

40

73

75

  1. The best WAC results for each classification scenario (binary or multi-class) are in italic. #F. and #P. C. stand for number of features and number of principal components, respectively