TY - GEN
T1 - Minimal Electrode EEG for BCI Emotion Detection
AU - Li, Yuxin
AU - Fang, Hao
AU - Liu, Wen
AU - Cheng, Chuantong
AU - Chen, Hongda
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Electroencephalography (EEG)-based emotion recognition is a potential research direction in the field of brain-computer interfaces (BCIs). However, its deployment on wearable devices still suffers from the challenges of low accuracy, heavy computation, and complex electrode placement. This study focuses on advancing the efficiency and cost-effectiveness of EEG-based BCIs for emotion recognition. Our approach begins with an investigation of electrode placement in relation to emotion detection, leveraging the SEED dataset to identify an optimal configuration that uses a minimal number of electrodes while maintaining high recognition accuracy. Employing a variety of machine learning and deep learning algorithms, we compare detection accuracy across different electrode combinations. Through these experiments and subsequent analysis, we identify an effective combination of two electrodes, T7 and T8, with the SVM method achieving an impressive 92.8 % accuracy. This finding laid the foundation for the design of our wearable, closed-loop BCI device with EEG-based emotion recognition capability.
AB - Electroencephalography (EEG)-based emotion recognition is a potential research direction in the field of brain-computer interfaces (BCIs). However, its deployment on wearable devices still suffers from the challenges of low accuracy, heavy computation, and complex electrode placement. This study focuses on advancing the efficiency and cost-effectiveness of EEG-based BCIs for emotion recognition. Our approach begins with an investigation of electrode placement in relation to emotion detection, leveraging the SEED dataset to identify an optimal configuration that uses a minimal number of electrodes while maintaining high recognition accuracy. Employing a variety of machine learning and deep learning algorithms, we compare detection accuracy across different electrode combinations. Through these experiments and subsequent analysis, we identify an effective combination of two electrodes, T7 and T8, with the SVM method achieving an impressive 92.8 % accuracy. This finding laid the foundation for the design of our wearable, closed-loop BCI device with EEG-based emotion recognition capability.
KW - Brain-Computer Interfaces
KW - Channel Selection
KW - Deep Learning
KW - Electroencephalography
KW - Emotion Recognition
KW - Machine Learning
UR - http://www.scopus.com/inward/record.url?scp=85192550530&partnerID=8YFLogxK
U2 - 10.1109/NNICE61279.2024.10499167
DO - 10.1109/NNICE61279.2024.10499167
M3 - Conference Proceeding
AN - SCOPUS:85192550530
T3 - 2024 4th International Conference on Neural Networks, Information and Communication Engineering, NNICE 2024
SP - 379
EP - 383
BT - 2024 4th International Conference on Neural Networks, Information and Communication Engineering, NNICE 2024
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 4th International Conference on Neural Networks, Information and Communication Engineering, NNICE 2024
Y2 - 19 January 2024 through 21 January 2024
ER -