Research on Emotional Classification of EEG Based on Convolutional Neural Network

Abstract

With the rapid development of machine learning technology, deep learning algorithms have been widely used in the processing of physiological signal. In this article, we used electroencephalography (EEG) signals based on Convolutional Neural Network (CNN) model in the deep learning algorithm to identify the positive and negative emotional states, and compare them with the support vector machine (SVM). By collecting the EEG signals of the subjects under different emotional stimuli states, the CNN and the SVM are used to identify the emotion data based on different feature transformations. The research results show that the average accuracy of using differential entropy (DE) features by SVM is 86.51%, which is better than the previous research on the same batch of datasets. At the same time, the classification effect of CNN is better than the traditional SVM (average classification accuracy is 86.90%), and its accuracy and stability have correspondingly better trends.