Emotion Assessment Based on EEG Brain Signals

Abstract

This paper presents an emotion assessment method that classifies five emotions (happy, sad, angry, fear, and disgust) using EEG brain signals. Public DEAP database is chosen for the proposed system evaluation, Fz channel electrode is selected for the feature extraction process. Then a Continuous Wavelet Transform (CWT) is used to extract the proposed Standard Deviation Vector (SDV) feature which describes brain voltage variation in both time and frequency domains. Finally, several machine learning classifiers are used for the classification stage. Experiment results show that the proposed SDV feature with SVM Classifier produce robust system with high accuracy result of about 91%.