基于同步性腦網(wǎng)絡(luò)的支持張量機(jī)情緒分類研究

黃麗亞; 蘇義博; 馬捃凱; 丁威威; 宋傳承

doi:10.11999/JEIT190882

基于同步性腦網(wǎng)絡(luò)的支持張量機(jī)情緒分類研究

doi: 10.11999/JEIT190882

1.
南京郵電大學(xué)電子與光學(xué)工程學(xué)院微電子學(xué)院南京 210023
2.
南京郵電大學(xué)貝爾英才學(xué)院南京 210023

基金項(xiàng)目: 國(guó)家自然科學(xué)基金(61977039)

詳細(xì)信息

作者簡(jiǎn)介:
黃麗亞：女，1972年生，教授，研究方向?yàn)槲锫?lián)網(wǎng)RFID技術(shù)、EDA技術(shù)以及通信網(wǎng)絡(luò)的QoS性能研究

蘇義博：男，1995年生，碩士生，研究方向?yàn)槟X電信號(hào)分析及嵌入式系統(tǒng)應(yīng)用

馬捃凱：男，1996年生，碩士生，研究方向?yàn)槟X電信號(hào)分析

丁威威：男，1996年生，碩士生，研究方向?yàn)榻?jīng)顱電刺激與人腦記憶力

通訊作者:
蘇義博　2524470353@qq.com

中圖分類號(hào): TN911.7; TP391.4
計(jì)量
- 文章訪問(wèn)數(shù): 2577
- HTML全文瀏覽量: 899
- PDF下載量: 111
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2019-11-04
- 修回日期: 2020-03-04
- 網(wǎng)絡(luò)出版日期: 2020-03-20
- 刊出日期: 2020-10-13

Research on Support Tensor Machine Based on Synchronous Brain Network for Emotion Classification

1.
School of Electronic and Optical Engineering & Microelectronics, Nanjing University of Posts and Telecommunications, Nanjing 210023, China
2.
Bell Honors Shool, Nanjing University of Posts and Telecommunications, Nanjing 210023, China

Funds: The National Natural Science Foundation of China (61977039)

摘要

摘要: 一直以來(lái)，情緒是心理學(xué)、教育學(xué)、信息科學(xué)等多個(gè)學(xué)科的研究熱點(diǎn)，腦電信號(hào)(EEG)因其客觀、不易偽裝的特點(diǎn)，在情緒識(shí)別領(lǐng)域受到廣泛關(guān)注。由于人類情緒是大腦多個(gè)腦區(qū)相互作用產(chǎn)生的，該文提出一種基于同步性腦網(wǎng)絡(luò)的支持張量機(jī)情緒分類算法(SBN-STM)，該算法采用相位鎖定值(PLV)構(gòu)建了同步性腦網(wǎng)絡(luò)，分析多導(dǎo)聯(lián)腦電信號(hào)之間的同步性和相關(guān)性，并生成2階張量序列作為訓(xùn)練集，運(yùn)用支持張量機(jī)(STM)模型實(shí)現(xiàn)正負(fù)情緒的二分類。該文基于DEAP腦電情緒數(shù)據(jù)庫(kù)，詳細(xì)分析了同步性腦網(wǎng)絡(luò)張量序列的選取方法，最佳張量序列窗口的大小和位置，解決了傳統(tǒng)情緒分類算法特征冗余的問(wèn)題，提高了模型訓(xùn)練速度。仿真實(shí)驗(yàn)表明，基于支持張量機(jī)的同步性腦網(wǎng)絡(luò)分類方法的情緒準(zhǔn)確率優(yōu)于支持向量機(jī)、C4.5決策樹(shù)、人工神經(jīng)網(wǎng)絡(luò)、K近鄰等以向量為特征的情緒分類模型。
- 情緒分類 /
- 同步性腦網(wǎng)絡(luò) /
- 支持張量機(jī) /
- 相位鎖定值
Abstract: Emotion has always been a research hot spot in many disciplines such as psychology, education, and information science. Electro EncephaloGram(EEG) signal has received extensive attention in the field of emotion recognition because of its objective and not easy to disguise. Since human emotions are generated by the interaction of multiple brain regions in the brain, an algorithm of Support Tensor Machine based on Synchronous Brain Network (SBN-STM) for emotion classification is proposed. The algorithm uses Phase Locking Value (PLV) to construct a synchronous brain network, in order to analyze the synchronization and correlation between multi-channel EEG signals, and generate a second-order tensor sequence as a training set. The Support Tensor Machine (STM) model can distinguish a two-category of positive and negative emotions. Based on the DEAP EEG emotion database, this paper analyzes the selection method of synchronic brain network tensor sequence, the research on the size and position of the optimal tensor sequence window solves the problem of traditional emotion classification algorithm which always exists feature redundancy, and improves the model training speed. The results show that the accuracy of the emotional classification method based on SBN-STM is better than support vector machine, C4.5 decision tree, artificial neural network, and K-nearest neighbor which using vectors as input feature.
- Emotion classification /
- Synchronous Brain Network (SBN) /
- Support Tensor Machine (STM) /
- Phase Locking Value (PLV)

HTML全文

圖 1 SBN-STM算法架構(gòu)圖

下載: 全尺寸圖片幻燈片

圖 2 張量序列窗口示意圖

下載: 全尺寸圖片幻燈片

圖 3 窗口半徑為1 s的張量序列示意圖

下載: 全尺寸圖片幻燈片

圖 4 32導(dǎo)聯(lián)位置示意圖^[24]

下載: 全尺寸圖片幻燈片

圖 5 各時(shí)刻的正向情緒PLV矩陣生成的灰度圖片

下載: 全尺寸圖片幻燈片

圖 6 各時(shí)刻腦網(wǎng)絡(luò)灰度圖及節(jié)點(diǎn)連接圖

下載: 全尺寸圖片幻燈片

圖 7 窗口不同中點(diǎn)位置準(zhǔn)確率比較

下載: 全尺寸圖片幻燈片

圖 8 窗口不同中點(diǎn)位置平均分類準(zhǔn)確率比較

下載: 全尺寸圖片幻燈片

圖 9 窗口不同中點(diǎn)位置分類準(zhǔn)確率盒須圖

下載: 全尺寸圖片幻燈片

圖 10 窗口在不同半徑下的分類準(zhǔn)確率比較

下載: 全尺寸圖片幻燈片

圖 11 窗口在不同半徑下的平均分類準(zhǔn)確率比較

下載: 全尺寸圖片幻燈片

圖 12 窗口在不同半徑下的準(zhǔn)確率盒須圖

下載: 全尺寸圖片幻燈片

表 1 各分類算法情緒二分類準(zhǔn)確率比較

數(shù)據(jù)集	分類算法	特征	特征類型	二分類準(zhǔn)確率(%)
DEAP數(shù)據(jù)庫(kù)	SBN-STM	同步性腦網(wǎng)絡(luò)	2階張量	78.30
	SVM^[2]	各頻段的功率譜密度	向量	73.30
	C4.5決策樹(shù)^[2]	各頻段的功率譜密度		72.50
	KNN^[3]	各小波頻段的能量、熵、統(tǒng)計(jì)特征		72.87
	ANN^[4]	α, β, θ 3個(gè)頻段上的雙譜		64.84
	LR^[25]	皮爾遜相關(guān)系數(shù)腦功能連接網(wǎng)絡(luò)	2階張量	70.22

下載: 導(dǎo)出CSV

參考文獻(xiàn)(25)

趙國(guó)朕, 宋金晶, 葛燕, 等. 基于生理大數(shù)據(jù)的情緒識(shí)別研究進(jìn)展[J]. 計(jì)算機(jī)研究與發(fā)展, 2016, 53(1): 80–92. doi: 10.7544/issn1000-1239.2016.20150636

ZHAO Guozhen, SONG Jinjing, GE Yan, et al. Advances in emotion recognition based on physiological big data[J]. Journal of Computer Research and Development, 2016, 53(1): 80–92. doi: 10.7544/issn1000-1239.2016.20150636

THAMMASAN N, MORIYAMA K, FUKUI K I, et al. Familiarity effects in EEG-based emotion recognition[J]. Brain Informatics, 2017, 4(1): 39–50. doi: 10.1007/s40708-016-0051-5

MERT A and AKAN A. Emotion recognition from EEG signals by using multivariate empirical mode decomposition[J]. Pattern Analysis and Applications, 2018, 21(1): 81–89. doi: 10.1007/s10044-016-0567-6

KUMAR N, KHAUND K, and HAZARIKA S M. Bispectral analysis of EEG for emotion recognition[J]. Procedia Computer Science, 2016, 84: 31–35. doi: 10.1016/j.procs.2016.04.062

PINCUS S M and VISCARELLO R R. Approximate entropy: A regularity measure for fetal heart rate analysis[J]. Obstetrics and Gynecology, 1992, 79(2): 249–255.

郝志峰, 謝蔚濤, 蔡瑞初, 等. 基于因果強(qiáng)度的時(shí)序因果關(guān)系發(fā)現(xiàn)算法[J]. 計(jì)算機(jī)工程與設(shè)計(jì), 2017, 38(1): 132–137. doi: 10.16208/j.issn1000-7024.2017.01.025

HAO Zhifeng, XIE Weitao, CAI Ruichu, et al. Causal inference on time series using causal strength[J]. Computer Engineering and Design, 2017, 38(1): 132–137. doi: 10.16208/j.issn1000-7024.2017.01.025

劉澄玉, 趙莉娜, 劉常春. 生理信號(hào)時(shí)間序列周期性和平穩(wěn)性對(duì)近似熵和樣本熵算法的影響分析[J]. 北京生物醫(yī)學(xué)工程, 2012, 31(2): 154–158, 163. doi: 10.3969/j.issn.1002-3208.2012.02.09

LIU Chengyu, ZHAO Li’na, and LIU Changchun. Influence analysis of physiological time-series periodicity and stability for approximate entropy and sample entropy[J]. Beijing Biomedical Engineering, 2012, 31(2): 154–158, 163. doi: 10.3969/j.issn.1002-3208.2012.02.09

ITO T, HEARNE L, MILL R, et al. Discovering the computational relevance of brain network organization[J]. Trends in Cognitive Sciences, 2020, 24(1): 25–38. doi: 10.1016/j.tics.2019.10.005

高佳, 王蔚. 基于稀疏貝葉斯網(wǎng)絡(luò)的情緒腦電的有效性腦網(wǎng)絡(luò)研究[J]. 生物醫(yī)學(xué)工程學(xué)雜志, 2015, 32(5): 945–951. doi: 10.7507/1001-5515.20150169

GAO Jia and WANG Wei. Research of effective network of emotion electroencephalogram based on sparse Bayesian network[J]. Journal of Biomedical Engineering, 2015, 32(5): 945–951. doi: 10.7507/1001-5515.20150169

武杰, 周春宇, 楊葉, 等. 基于組獨(dú)立成分分析方法的情緒刺激對(duì)腦部激活區(qū)域的研究[J]. 生物醫(yī)學(xué)工程學(xué)進(jìn)展, 2018, 39(3): 125–129. doi: 10.3969/j.issn.1674-1242.2018.03.001

WU Jie, ZHOU Chunyu, YANG Ye, et al. Research on the brain activation region from emotion stimulation based on the group ICA method[J]. Progress in Biomedical Engineering, 2018, 39(3): 125–129. doi: 10.3969/j.issn.1674-1242.2018.03.001

MAHYARI A G, ZOLTOWSKI D M, BERNAT E M, et al. A tensor decomposition-based approach for detecting dynamic network states from EEG[J]. IEEE Transactions on Biomedical Engineering, 2017, 64(1): 225–237. doi: 10.1109/TBME.2016.2553960

金廣智, 石林鎖, 崔智高, 等. 結(jié)合GLCM與三階張量建模的在線目標(biāo)跟蹤[J]. 電子與信息學(xué)報(bào), 2016, 38(7): 1609–1615. doi: 10.11999/JEIT151108

JIN Guangzhi, SHI Linsuo, CUI Zhigao, et al. Online object tracking based on gray-level co-occurrence matrix and third-order tensor[J]. Journal of Electronics &Information Technology, 2016, 38(7): 1609–1615. doi: 10.11999/JEIT151108

LIU Chuanwei, FU Yunfa, YANG Jun, et al. Discrimination of motor imagery patterns by electroencephalogram phase synchronization combined with frequency band energy[J]. IEEE/CAA Journal of Automatica Sinica, 2017, 4(3): 551–557. doi: 10.1109/JAS.2016.7510121

CHEN Yanyan, WANG Kuaini, and ZHONG Ping. One-class support tensor machine[J]. Knowledge-Based Systems, 2016, 96: 14–28. doi: 10.1016/j.knosys.2016.01.007

DIAN Renwei, LI Shutao, and FANG Leyuan. Learning a low tensor-train rank representation for hyperspectral image super-resolution[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(9): 2672–2683. doi: 10.1109/TNNLS.2018.2885616

CAI Deng, HE Xiaofei, WEN Jirong, et al. Support tensor machines for text categorization[R]. UIUCDCS-R-2006-2714, 2006: 2222–6990.

CHEN Yuee and REN Baili. Research on large scale data set processing based on SVM[J]. Advanced Materials Research, 2011, 216: 738–741. doi: 10.4028/www.scientific.net/amr.216.738

ZHOU Bingyin, SONG Biao, HASSAN M M, et al. Multilinear rank support tensor machine for crowd density estimation[J]. Engineering Applications of Artificial Intelligence, 2018, 72: 382–392. doi: 10.1016/j.engappai.2018.04.011

CHEN Yanyan, LU Liyun, and ZHONG Ping. One-class support higher order tensor machine classifier[J]. Applied Intelligence, 2017, 47(4): 1022–1030. doi: 10.1007/s10489-017-0945-9

LI Zhibao, DAI Yuhong, and GAO Huan. Alternating projection method for a class of tensor equations[J]. Journal of Computational and Applied Mathematics, 2019, 346: 490–504. doi: 10.1016/j.cam.2018.07.013

馮翔, 陳志坤, 趙宜楠, 等. 基于聯(lián)合優(yōu)化松弛交替投影的組網(wǎng)雷達(dá)恒模波形設(shè)計(jì)[J]. 電子與信息學(xué)報(bào), 2016, 38(7): 1745–1751. doi: 10.11999/JEIT151152

FENG Xiang, CHEN Zhikun, ZHAO Yi’nan, et al. Unimodular waveforms design for netted radar system via joint optimization relaxed alternating projection[J]. Journal of Electronics &Information Technology, 2016, 38(7): 1745–1751. doi: 10.11999/JEIT151152

SHI Haifa, ZHAO Xinbin, and JING Ling. Tensor distance based least square twin support tensor machine[J]. Applied Mechanics and Materials, 2014, 668. doi: 10.4028/www.scientific.net/amm.668-669.1170

CYGANEK B and WOZNIAK M. Efficient computation of the tensor chordal kernels[J]. Procedia Computer Science, 2016, 80: 1702–1711. doi: 10.1016/j.procs.2016.05.511

KOELSTRA S, MüHL C, SOLEYMANI M, et al. DEAP: A database for emotion analysis; Using physiological signals[J]. IEEE Transactions on Affective Computing, 2012, 1(3): 18–31. doi: 10.1109/t-affc.2011.15

趙少楷. 基于EEG腦網(wǎng)絡(luò)特征的情緒識(shí)別研究[D]. [碩士論文], 杭州電子科技大學(xué), 2018.

ZHAO Shaokai. The research of emotion recognition based on features of brain networks[D]. [Master dissertation], Hangzhou Dianzi University, 2018.

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問(wèn)統(tǒng)計(jì)