Lightweight Building of an Electroencephalogram-Based Emotion Detection System

Area of Science:

• Neuroscience
• Computer Science
• Artificial Intelligence

Background:

• Brain-computer interface (BCI) technology enables direct communication between the brain and external devices.
• Electroencephalogram (EEG) signals are used for monitoring brain activity and detecting human emotions.
• Recent advancements focus on developing efficient EEG-based emotion detection systems for practical applications.

Purpose of the Study:

• To investigate the feasibility of a lightweight EEG-based emotion detection system using spiking neural networks.
• To reduce the reliance on large EEG datasets and complex feature extraction methods.
• To maintain high accuracy in emotion detection with minimal data.

Main Methods:

• Utilized a spiking neural network (NeuCube) for emotion detection.
• Employed a reduced version of the DEAP dataset.
• Avoided traditional feature extraction methods.

Main Results:

• Successfully detected valence emotion levels using only 60 EEG samples.
• Achieved an accuracy of 84.62% in emotion detection.
• Demonstrated comparable accuracy to previous studies with significantly less data and complexity.

Conclusions:

• Spiking neural networks offer a viable approach for building efficient EEG-based emotion detection systems.
• The proposed method is lightweight and effective, requiring minimal data and no feature extraction.
• This research supports the development of practical BCI applications for human-machine interaction.