Basic Emotion Recognition System for Persons with Amyotrophic Lateral Sclerosis Using Electroencephalography

Abstract

Amyotrophic Lateral Sclerosis (ALS) is a progressive neurodegenerative disorder that causes the death of motor neurons making the person unable to speak and move. An EEG device will be used to record the brain activity of the user especially the emotions which were classified into three: the happy, sad and angry. The brain signals will be stored to be used as data. The method consists of pre-processing, amplification of the raw EEG signal and filtering. The filtered EEG signals will then undergo feature extraction to extract and determine the fundamental frequency components of the signal. The extracted frequency parameters will be the input for signal classification that will differentiate the three emotions. This study shows that the system achieved a 93.33% accuracy for happy emotion; 86.66% for sad; and 83.33% for angry detection, certainly proving to have a good score and beneficial utility for the recognition of emotion through EEG.