Skip to main content

Table 2 Comparison of different schemes of linear classifier

From: Emotion classification based on brain wave: a survey

SchemesPreprocessingFeature extractionFeature smoothingClassificationEmotion statesAccuracy
Method by Li et al.FTCSP liner-SVMHappiness and sadness93.5%
Method by Murugappan et al.Surface Laplacian filtering
Zero mean unit variance
Wavelet transform KNN and LDADisgust, happy, surprise, fear and neutralKNN: 77.68%
LDA: 73.5%
Method by Wang et al. Wavelet transform, PCA, LDA, CFSLDSliner-SVMNegative and positive87.53%
Method by Petrantonakis et al. Statistical values, wavelet transform and HOC QDA, KNN, MD and SVMsHappiness, surprise, anger, fear, disgust and sadnessQDA: 62.3%
SVMs: 83.33%
MD: 44.90%
KNN: 34.60%
Method by Duan et al. DE, DASM, RASM and ESLDS
PCA and MRMR
liner-SVM and kNNNegative and positiveliner-SVM: 74.10%
kNN: 69.24%