Differentiating Active and Natural Emotional Expressions Using EEG Signals with ML

This research essentially reviews the discrimination of emotional identification states with the valance (or) Arousal model by electroencephalography (EEG) signals. Frequency bands are used to categorize EEG signals named gamma, alpha, beta, and theta by power spectral density (PSD). Spectral characteristics are unchained from gamma, alpha, beta, and theta frequency bands. Extracted elements have some extensions developed using spectral power density, which makes characteristics unrelated. Emotive conditions are categorized based on K-nearest neighbor (KNN), support vector machine (SVM), and artificial neural network (ANN). Ten EEG channels were extracted, and the cross-validated support vector machine (SVM) with radial basic function (RBF) kernel was used. In this study, we increase the emotional identification accuracy by utilizing SVM classifier. Emotion identification is the technique of identifying human emotions. Individuals differ widely in their capacity to identify the emotions of other individuals. The topic of research on using technology to help people recognize their emotions is very new. In general, technology works best when various modalities are used in conjunction. When studying about automated emotion recognition, it’s important to remember that there are a variety of sources of information about what the true emotion is.