EEG based emotion recognition using SVM and PSO

Machine learning has fueled real breakthroughs in affective computing in making the machines more emphatic to the user. This emotion recognition capability of machines enables them to act according to the observed mental state. Human feelings and emotions are triggered by stimuli which are external or internal and manifest themselves in the form of pulse rate, tone, facial expressions and many more. In this paper we classify human emotions using EEG signals into four discrete states, namely happy, sad, angry and relaxed. The preprocessed signals from the DEAP database is used and spectral and statistical features are extracted by discrete wavelet transform. These features are classified using a SVM classifier and the performance of the classifier is optimized using the PSO algorithm. An overall emotional accuracy of 80.625% was obtained for a combination of 32 electrodes with a valence and arousal accuracy of 86.25% and 88.125%.