Automation of Emotion Quadrant Identification by Using Second Order Difference Plots and Support Vector Machines

EEG can reveal the real internal emotion of the subject, as it is a non-invasive way of capturing the brain waves and can't be affected by pretension or denial. This has made EEG a reliable source for research on emotion in current times, as it cannot be disguised. It captures the brain activations, mapping the brain states representing different emotional states directly [1]. Emotion recognition from EEG signals is a very cost-effective method to monitor the general wellbeing of individuals, employees of an organization or to cater to patients of mental health. Such a dataset is DEAP - Database for Emotion Analysis using Physiological signals, which is available online for academic research purposes [2]. In DEAP emotional dataset, brain signals of 32 volunteers, captured as they viewed 40 music videos of 1-minute duration each, are categorized on the quadrant of valence, arousal, dominance and liking, which signifies how they are associated with different emotions. The overview of the dataset used for experimentation is as shown in Figure 1.