EEG-Based Emotion Recognition with Deep Convolution Neural Network

Emotions are closely related to people's work and life. Emotional analysis and recognition is not only an urgent need to solve certain mental illnesses, but also has broad application prospects in the fields of human-computer interaction, entertainment and medical care. Therefore, it is of great value to classify emotional EEG signals. This paper introduces CNN(Convolutional Neural Networks)into the process of emotional EEG recognition. The innovation of this method is to adjustthe convolution kernel of the CNN to adapt to the input of EEG signals. The classification accuracy of 0.8579 is achieved in the three-classification emotional EEG signal.