A Hierarchical Bidirectional GRU Model With Attention for EEG-Based Emotion Classification

被引:137
作者
Chen, J. X. [1 ,2 ]
Jiang, D. M. [1 ]
Zhang, N. [1 ]
机构
[1] Northwestern Polytech Univ, Sch Comp Sci & Engn, Xian 710072, Shaanxi, Peoples R China
[2] Shaanxi Univ Sci & Technol, Dept Elect Informat & Artificial Intelligence, Xian 710021, Shaanxi, Peoples R China
基金
中国国家自然科学基金;
关键词
Hierarchical; bidirectional GRU; attention; EEG; emotion classification; RECOGNITION;
D O I
10.1109/ACCESS.2019.2936817
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we propose a hierarchical bidirectional Gated Recurrent Unit (GRU) network with attention for human emotion classification from continues electroencephalogram (EEG) signals. The structure of the model mirrors the hierarchical structure of EEG signals, and the attention mechanism is used at two levels of EEG samples and epochs. By paying different levels of attention to content with different importance, the model can learn more significant feature representation of EEG sequence which highlights the contribution of important samples and epochs to its emotional categories. We conduct the cross-subject emotion classification experiments on DEAP data set to evaluate the model performance. The experimental results show that in valence and arousal dimensions, our model on 1-s segmented EEG sequences outperforms the best deep baseline LSTM model by 4.2% and 4.6%, and outperforms the best shallow baseline model by 11.7% and 12% respectively. Moreover, with increase of the epoch's length of EEG sequences, our model shows more robust classification performance than baseline models, which demonstrates that the proposed model can effectively reduce the impact of long-term non-stationarity of EEG sequences and improve the accuracy and robustness of EEG-based emotion classification.
引用
收藏
页码:118530 / 118540
页数:11
相关论文
共 36 条
[1]  
Alhagry S, 2017, INT J ADV COMPUT SC, V8, P355, DOI 10.14569/IJACSA.2017.081046
[2]  
[Anonymous], 2017, arXiv
[3]  
[Anonymous], 2015, P 2015 C EMP METH NA
[4]  
Bahdanau D., 2019, ARXIV14090473
[5]  
Bashivan P., 2015, Learning Representations from EEG with Deep Recurrent-Convolutional Neural Networks
[6]   Driver Fatigue Classification With Independent Component by Entropy Rate Bound Minimization Analysis in an EEG-Based System [J].
Chai, Rifai ;
Naik, Ganesh R. ;
Tuan Nghia Nguyen ;
Ling, Sai Ho ;
Tran, Yvonne ;
Craig, Ashley ;
Nguyen, Hung T. .
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2017, 21 (03) :715-724
[7]   Accurate EEG-Based Emotion Recognition on Combined Features Using Deep Convolutional Neural Networks [J].
Chen, J. X. ;
Zhang, P. W. ;
Mao, Z. J. ;
Huang, Y. F. ;
Jiang, D. M. ;
Zhang, Andy N. .
IEEE ACCESS, 2019, 7 :44317-44328
[8]   Feature Selection of Deep Learning Models for EEG-Based RSVP Target Detection [J].
Chen, Jingxia ;
Mao, Zijing ;
Zheng, Ru ;
Huang, Yufei ;
He, Lifeng .
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2019, E102D (04) :836-844
[9]   A Common Spatial Pattern and Wavelet Packet Decomposition Combined Method for EEG-Based Emotion Recognition [J].
Chen, Jingxia ;
Jiang, Dongmei ;
Zhang, Yanning .
JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS, 2019, 23 (02) :274-281
[10]  
[陈继文 Chen Jiwen], 2019, [现代制造工程, Modern Manufacturing Engineering], P1