Framework for the Classification of Emotions in People With Visual Disabilities Through Brain Signals

Nowadays, the recognition of emotions in people with sensory disabilities still represents a challenge due to difficulty generalizing and modeling set brain signals. In recent years, technology that has been used study person’s behavior based on signals is brain–computer interface (BCI). Although previous works have already proposed classification using machine learning techniques, model visual not yet evaluated. Consequently, this work, authors present twofold framework focused disabilities. Firstly, auditory stimuli used, component acquisition extraction defined. Secondly, analysis techniques for developed, models Based results, algorithm best performance validation random forest (RF), an accuracy 85 88% negative positive emotions, respectively. According able classify but experimentation performed also shows depends number features dataset quality Electroencephalogram (EEG) determining factor.