Classification of emotions based on ERP feature extraction

Emotions are the feelings that represent the personality of any individual. Thus predicting emotions become necessary to understand the behavior of humans. Emotions can be predicted from gestures, sound processing but emotion recognition using EEG signals is very powerful method to know the internal state of mind accurately. This paper describes the acquisition of EEG signals on frontal electrodes such as F3, F4 and FZ from five subjects for classification of emotions into two classes. The emotions were induced by showing images from International Affective Picture System (IAPS) dataset to the subjects. The event related potential (ERP) features were determined from the processed EEG signals for every class of emotions. The classification was performed using LIBSVM classifier with 3 fold cross validation and RBF kernel to classify emotions into two classes along the arousal axis. It was found that accuracy remained consistently high on F4 electrode. An accuracy of 79.16% was obtained on F4 electrode, 76.19% on F3 electrode and 73.07% on FZ electrode when classifying emotions subject wise.