Human Emotion Classification from Brain EEG Signal Using Multimodal Approach of Classifier

Abstract

To deeply understand the brain response under different emotional states can fundamentally advance the computational models for emotion recognition. Various psychophysiology studies have demonstrated the correlations between human emotions and EEG signals. With the quick development of wearable devices and dry electrode techniques it is now possible to implement EEG-based emotion recognition from laboratories to real-world applications. In this paper we have developed EEG-based emotion recognition models for three emotions: positive, neutral and negative. Extracted features are downloaded from seed database to test a classification method. Gamma band is selected as it relates to emotional states more closely than other frequency bands. The linear dynamical system (LDS) is used to smooth the features before classification. The classification accuracy of the proposed system using DE, ASM, DASM, RASM is 97.33, 89.33 and 98.37 for SVM (linear), SVM (rbf sigma value 6) and KNN(n value 3) respectively.