EEG-Based Emotion Recognition Using SVM

Emotional recognition and detection has become one of the fastest emergent areas which has encouraged the researchers to do more research in this area. In this paper, we described an EEG-based emotion recognition method which identifies the human emotional states. The proposed method used the DEAP database and developed a supervised machine learning algorithms to recognize the emotional from EEG signal. The Welch’s transformer is used to preprocess the EEG signal to obtain the alpha, delta, theta, beta waves and gamma waves. Others attributes such as mean, variance, standard deviation, kurtosis and skewness are also collected. After using SVM and XGbooster supervised machine learning algorithms, appropriate emotional state has been detected. The proposed method achieved the 96.2% response rate for happy, 94.8% for neutral, 93% for sad and angry. The proposed method has been compared with existing method.