Table 1: EEG-based emotion recognition researches.

References YearParticipantEmotionStimulusFeatureTemporal windowClassifierResultReal time

[10]20064
subject-dependent
3 arousal classesPicture PSDNB58%No
[11]200826
subject-independent
4 classes (joy, anger, sadness, and pleasure)Music ASM1 sSVM92.73%No
[20]200910
subject-dependent
2 valence classesPicture CSP3 sSVM93.5%No
[21]200910
3 arousal classesRecall PSD0.5 sSVM63%No
[22]20091
subject-dependent
3 classes (positively excited, negatively excited, and calm)Picture statistical featuresQDA66.66%No
[23]20093
subject-dependent
10 classesSelf-elicitedPSD1 sKNN39.97–66.74%No
[24]201026
subject-independent
4 classes (joy, anger, sadness, and pleasure)Music ASM1 sSVM82.29%No
[25]20106
subject-dependent
2 valence classes
2 arousal classes
Music videoPSDSVM58.8% (valence)
55.7% (arousal)
No
[26]201026
subject-dependent
4 classes (calm, happy, sad, and fear)Picture and musicSOM2 sKNN84.5%No
[28]201015
2 classes (calm-neutral and negatively excited)Picture HOS2 sSVM82%No
[29]201012
subject-dependent
2 valence classes
2 arousal classes
SoundFDthresholdYes
[27]201120
5 classes (happy, disgust, surprise, fear, and neutral) video clipEntropyKNN 83.04%No
[31]20116
subject-dependent
2 valence classesMovie clipPSD1 sSVM87.53%No
[32]201120
subject-independent
3 classes (boredom, engagement, and anxiety)Game PSDLDA56%No
[33]20115
subject-dependent
4 classes (joy, relax, sad, and fear)MoviePSD1 sSVM66.51%No
[34]201111
3 valence classesPicture ASM4 sKNN82%No
[30]201227
subject-independent
3 valence classes
3 arousal classes
Video PSD and ASMSVM57.0% (valence)
52.4% (arousal)
No
[35]201232
2 valence classes
2 arousal classes
Music videoPSD and ASMNB57.6% (valence)
62.0% (arousal)
No
[36]201220
subject-dependent
5 classes (happy, angry, sad, relaxed, and neutral)Picture FDSVM70.5%Yes
[37]20125
subject-dependent
3 classes (positively excited, negatively excited, and calm)Picture HOCKNN90.77%No
[38]20124
2 valence classes
2 arousal classes
Video clipASP66.05% (valence)
82.46% (arousal)
No
[39]201232
2 classes (stress and calm)Music videoPSDKNN70.1%No
[40]201236
3 classesMusic videoPSDANNYes
[41]201311
subject-independent
2 valence classesPicture PSD4 sSVM85.41%No

The feature, temporal window, and classifier shown in this table are the sets giving the best accuracy of each research.
Feature: Power Spectral Density (PSD), Spectral Power Asymmetry (ASM), Common Spatial Pattern (CSP), Higher Order Crossings (HOC), Self-Organizing Map (SOM), Higher Order Spectra (HOS), Fractal Dimension (FD), and Asymmetric Spatial Pattern (ASP).
Classifier: Support Vector Machine (SVM), Naïve Bayes (NB), Quadratic Discriminant Analysis (QDA), K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Multilayer Perceptron (MLP), and Artificial Neural Network (ANN).