MGGL: A multi-grained graph learning model for EEG emotion recognition

Key Points

• Emotion recognition accuracy improves with the proposed graph learning model, demonstrating effective classification results.
• Achieving a performance boost of over 10% in classification tasks compared to traditional methods, using EEG signals.
• Assessment using neural networks for feature extraction showcases how multimodal data enhances recognition.
• This approach highlights the necessity for advanced emotion recognition systems in emotional and mental health monitoring.