脑电时间效应对情绪识别模型的影响及解决措施研究

doi:10.3969/j.issn.0258-8021. 2017. 01.005

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (3069 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract There are numerous studies measuring the brain emotional status by analyzing EEGs under the emotional stimuli that have occurred, however, in practical application, an important but unresolved question is the extent to which the emotion model may generalize over time, since people could have a different expression of the same physiological signal on different days even when he experiences the same emotion. This paper attempted to add multiple days to the training set in purpose to weaken the impact of day-effect, and then to improve the generalization of the classifier. Eight subjects participated in this experiment, in which movie clips were presented to evoke the subjects’ three emotional states of neutral, positive and negative. Moreover, EEG was recorded 5 times within one month for each subjects. Support vector machine (SVM) was used to obtain the 3-class classification rates in all the collecting conditions including1-day collection, 2-day collection, 3-day collectionand 4-day collection. N-day collection represented the case in which data from N days were sent to train the SVM and the remaining (5-N) days were used to form the testing set. Results showed that the accuracy was increased with the number of days in the training set for most of the subjects. Compared with 1-day collection, the increasing rates of the accuracies were 6.45%(P=0.006), 10.48%(P=0.000), and 14.40%(P=0.000)in 2-day, 3-day and 4-day collections. These results suggested that adding data from more days to the training set could improve the performance and generalization of an emotion classifier. Though it is still a big challenge in EEG-based emotion recognition, these results provided a promising solution and take EEG-based model one step closer to being able to discriminate emotions in practical application.

Key words： electroencephalography (EEG) emotion recognition day-effect time robustness support vector machine (SVM)

Received: 27 April 2016

PACS:

R318

About author:: (Senior member, Chinese Society of Biomedical Engineering)

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Liu Shuang
	Tong Jingjing
	Guo Dongyue
	Ke Yufeng
	Meng Jiayuan
	Chen Long
	Xu Minpeng
	Qi Hongzhi
	Ming Dong

Cite this article:

Liu Shuang,Tong Jingjing,Guo Dongyue, et al. Study on Time-Effect on EEG-BasedEmotion Recognition Model and Countermeasures[J]. Chinese Journal of Biomedical Engineering, 2017, 36(1): 39-45.

URL:

http://cjbme.csbme.org/EN/10.3969/j.issn.0258-8021. 2017. 01.005 OR http://cjbme.csbme.org/EN/Y2017/V36/I1/39

[1] Blanchette I, Richards A. The influence of affect on higher level cognition: A review of research on interpretation, judgement, decision making and reasoning [J]. Cognition & Emotion, 2010, 24(4): 561-595.
[2] Picard RW, Vyzas E, Healey J. Toward machine emotional intelligence: Analysis of affective physiological state [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(10): 1175-1191.
[3] Liu S, Zhang D, Xu M, et al. Randomly dividing homologous samples leads to overinflated accuracies for emotion recognition [J]. International Journal of Psychophysiology, 2015, 96(1): 29-37.
[4] Chueh TH, Chen TB, Lu HHS, et al. Statistical prediction of emotional states by physiological signals with Manova and machine learning [J]. International Journal of Pattern Recognition and Artificial Intelligence, 2012, 26(4): 1025-1035.
[5] Verma GK, Tiwary US. Multimodal fusion framework: A multiresolution approach for emotion classification and recognition from physiological signals [J]. NeuroImage, 2014, 102:162-172.
[6] Wang XW, Nie D, Lu BL. Emotional state classification from EEG data using machine learning approach [J]. Neurocomputing, 2014, 129:94-106.
[7] Khalili, Z, Moradi, M.H. Emotion detection using brain and peripheral signals[C]// Biomedical Engineering Conference. Piscataway: IEEE, 2009:737-738.
[8] Cohen B, Bravo-Fernandez E, Sances Jr A. Automated electroencephalographic analysis as a prognostic indicator in stroke [J]. Medical and Biological Engineering and Computing, 1977, 15(4): 431-437.
[9] 万柏坤, 朱欣, 杨春梅, 等. ICA去除EEG中眼动伪差和工频干扰方法研究 [J]. 电子学报, 2003, 31(10): 1571-1574.
[10] 陈海英. AR模型功率谱估计常用算法的性能比较 [J]. 漳州师范学院学报(自然科学版), 2009, 22(1): 48-52.
[11] Hidalgo-Mu Oz AR, L Pez MM, Santos IM, et al. Application of SVM-RFE on EEG signals for detecting the most relevant scalp regions linked to affective valence processing [J]. Expert Systems with Applications, 2013, 40(6): 2102-2108.
[12] Chang CC, Lin CJ. LIBSVM: a Library for Support Vector Machines [J]. Acm Transactions on Intelligent Systems & Technology, 2006, 2(3): 389-396.
[13] 张文彤. SPSS统计分析基础教程 [M]. 北京:高等教育出版社, 2011.
[14] Brown L, Grundlehner B, Penders J. Towards wireless emotional valence detection from EEG.[C]// International Conference of the IEEE Engineering in Medicine & Biology Society.Piscataway: IEEE, 2011:2188-2191.
[15] Shuang L, Di Z, Minpeng X, et al. Randomly dividing homologous samples leads to overinflated accuracies for emotion recognition [J]. International Journal of Psychophysiology, 2015, 96(1): 29-37.