Human Emotion Classification from EEG Signals Using Multiwavelet Transform

In this paper, we propose new features based on multiwavelet transform for classification of human emotions from electroencephalogram (EEG) signals. The EEG signal measures electrical activity of the brain, which contains lot of information related to emotional states. The sub-signals obtained by multiwavelet decomposition of EEG signals are plotted in a 3-D phase space diagram using phase space reconstruction (PSR). The mean and standard deviation of Euclidian distances are computed from 3-D phase space diagram. These features have been used as input features set for multiclass least squares support vector machines (MC-LS-SVM) together with the radial basis function (RBF), Mexican hat wavelet and Morlet wavelet kernel functions for classification of emotions. The proposed features based on multiwavelet transform of EEG signals with Morlet wavelet kernel function of MC-LS-SVM have provided better classification accuracy for classification of emotions.