Skip to main content

Table 7 Results of all binary classification models with their respective cross-validation and their accuracy of the balanced database using ADASYN

From: A comparative study of CNN-capsule-net, CNN-transformer encoder, and Traditional machine learning algorithms to classify epileptic seizure

Ā Ā Ā 

Epileptic

Without Epileptic

Ā 

Accuracy [%]

Cross Validation [%]

Precision [%]

Precision [%]

ETC

97.87

98.50 Ā± 1.0

98.90

96.80

RFC

97.04

96.00 Ā± 2.0

99.00

95.20

GB

98.18

98.20 Ā± 1.0

97.50

98.80

DTC

89.96

91.00 Ā± 3.0

89.50

90.40

MLP

98.60

98.70 Ā± 0.1

99.60

97.60

KNN

99.81

1.000 Ā± 1.5

99.60

99.60

SGD

61.44

62.50 Ā± 1.0

58.00

71.90

SVM

98.63

98.80 Ā± 0.5

99.70

97.60

CNN+Fully

99.40

99.50 Ā± 1

99.50

99.30

CNN+Capsule-Net

98.22

98.50 Ā± 0.4

98.40

98.10

CNN+TF

94.42

92.80 Ā± 2.0

90.70

99.00

CNN+Tf+FULLY

97.83

96.50 Ā± 1.0

99.9

97.90

CNN+TF+Capsule-Net

98.91

98.20 Ā± 1.0

99.10

98.70