GSCNN: Gender-Sensitive EEG Emotion Recognition Using Convolutional Neural Network

Emotion recognition plays an important role in the field of human-computer interaction. The previous studies have demonstrated the existence of gender differences in EEG signals, which leads to undesirable EEG emotion recognition in cross-subject tasks. However, there is no EEG emotion recognition model that involves gender factors. In this paper, a gender-sensitive convolutional neural network is proposed to enhance the performance of EEG emotion recognition. Specifically, we design a gender information extractor to help the network learn the individual difference during classification. Meanwhile, we also design an emotion information extractor to assign different weights for key frequency bands according to their classification performance. The performance of the proposed method is evaluated on two publicly available EEG emotion recognition datasets, SEED and SEED-IV. The experimental results show that our model can achieve higher recognition accuracies of 85.6% and 63.3% on the two datasets, respectively.