Skip to main content

Table 3 Comparison of different schemes of nonlinear classifiers

From: Emotion classification based on brain wave: a survey

SchemesPreprocessingFeature extractionFeature smoothingClassificationEmotion statesAccuracy
Method by Liu et al.   FDSad, frustrated, fear, satisfied, pleasant and happy 
Method by Liu et al. ResNets, LFCC KNN, SVM, LR, RF, NB, DT and FCAnger, joy, sadness and pleasureKNN: 89.72%
Method by Zheng et al. DE, DASM, RASM DBN, SVM, LR and KNNPositive, neutral and negativeDBN: 86.08%
SVM: 83.99%
LR: 82.70%
KNN: 72.60%
Method by Dan Nie et al. FFTLDSSVMNegative and positiveSVM: 87.53
Method by Zheng et al. PSD, DE, DASM, RASM, ASM and DCAUMRMRKNN, LR, SVM and GELMNegative, positive and neutralKNN: 70.43%
LR: 84.08%
SVM: 78.21
GELM: 91.07%