Emotion Recognition Using Brain Signals

James-Lange defined emotion as "A positive or negative experience that is associated with a particular pattern of physiological activity". Emotions produce different psycho physiological signal which is either stimulated by conscious or unconscious profundity of a situation or an object. Emotion has a key role in human communication as well as in modern BCI (brain-computer interaction) systems. Most of the contemporary BCI systems are lacking emotional intelligence and also they are not able to detect human emotional states for proper execution of action. Electroencephalogram (EEG) is one of the common methods to acquire brain signals for brain computer interface (BCI). Most of the works for emotion recognition using EEG data are complex and with relatively average performance obtained. Hence, there is still a lot of space for developing a better performing system in emotion recognition. In the present work, a wavelet based better performing feature extraction algorithm is proposed. Further this algorithm is tested for different classifiers namely k Nearest Neighbor (KNN), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM). The average sensitivity obtained for the present work was 92.97% with an accuracy of 91.67%, which is better result than the previous work.