Table 5 Subject-independent classification accuracy and details of the proposed method and hybrid method and CNN classifier.

From: Accurate emotion recognition using Bayesian model based EEG sources as dynamic graph convolutional neural network nodes

EEG source localization method

The type of classifier

SEED dataset (%)

Recorded dataset (%)

DEAP dataset (%)

sLORETA

CNN

95.56

93.75

95.18

sLORETA

DGCNN

97.78

97.50

97.65

Bernoulli–Laplace-based Bayesian model

CNN

96.67

95.31

96.48

Bernoulli–Laplace-based Bayesian model

DGCNN

98.51

97.91

98.31

Back to article page