EEG-Based Emotion Recognition with Consideration of Individual Difference.

Area of Science:

• Neuroscience
• Machine Learning
• Affective Computing

Background:

• Electroencephalograms (EEGs) are utilized for emotion recognition via trained models.
• Individual differences in emotional responses to stimuli can reduce model accuracy.
• Existing models often overlook participant-specific variations in emotion processing.

Purpose of the Study:

• To propose a brain-based emotion recognition model that addresses individual differences.
• To enhance the accuracy and robustness of EEG-based emotion classification.
• To improve practical applications of emotion recognition technology.

Main Methods:

• Developed a model with an emotion classification module and an individual difference module (IDM).
• The IDM incorporates personalized adjustments for participant-specific variations.
• Utilized spatial and temporal feature extraction from EEG data.

Main Results:

• Achieved a maximum accuracy of 96.43% for binary classification on DEAP dataset.
• Demonstrated superior performance in scenarios with significant individual differences, reaching 98.92% accuracy.
• The proposed model effectively mitigates the impact of inter-individual variability.

Conclusions:

• The proposed brain-based model significantly enhances EEG emotion recognition by considering individual differences.
• This approach offers improved accuracy and reliability for diverse populations.
• The model shows promise for more effective real-world emotion recognition applications.