Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data

Hongquan Qu; Hanwen Dong; Liping Pang

doi:10.1007/978-981-19-4786-5_3

Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data

Hongquan Qu
, Hanwen Dong
, Liping Pang^*

^*Corresponding author for this work

School of Aeronautic Science and Engineering

North China University of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Due to the non-stationary randomness of the EEG signals, which results in a trained classifier with very low generalization ability. In order to improve the cross-session classification accuracy of Mental Workload (MW), a domain adaptive classification method is studied based on Transfer Component Analysis (TCA) using EEG Independent Components (ICs) in this paper. First step is to transform the EEG data in source and target domains in a same way. The TCA algorithm is used to build the transformation ϕ in order to minimize their Maximum Mean Discrepancy (MMD) distance between the source domain and the target domain. Second step is to train a MW classifier using the SVM and the transformed EEG data in the source domain. The target domain data are used for test. The presented method is compared with Subspace Alignment (SA). EEG data in the source and target domains are measured 24 h apart. A-distance is calculated to reveal the distribution difference between the transformed source and the target domains. The results show that the classification accuracy of the proposed method can be improved by 15.11% compared with SA. The A-distance of TCA is significantly lower than that of SA.

Original language	English
Title of host publication	Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE
Editors	Shengzhao Long, Balbir S. Dhillon
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	17-23
Number of pages	7
ISBN (Print)	9789811947858
DOIs	https://doi.org/10.1007/978-981-19-4786-5_3
State	Published - 2023
Event	22nd International Conference on Man-Machine-Environment System Engineering, MMESE 2022 - Beijing, China Duration: 21 Oct 2022 → 23 Oct 2022

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	941 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	22nd International Conference on Man-Machine-Environment System Engineering, MMESE 2022
Country/Territory	China
City	Beijing
Period	21/10/22 → 23/10/22

Keywords

Classification of mental workload
EEG
Independent Component Analysis (ICA)
Subspace Alignment (SA)
Transfer Component Analysis (TCA)

Access to Document

10.1007/978-981-19-4786-5_3

Cite this

Qu, H., Dong, H., & Pang, L. (2023). Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data. In S. Long, & B. S. Dhillon (Eds.), Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE (pp. 17-23). (Lecture Notes in Electrical Engineering; Vol. 941 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-19-4786-5_3

Qu, Hongquan ; Dong, Hanwen ; Pang, Liping. / Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data. Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE. editor / Shengzhao Long ; Balbir S. Dhillon. Springer Science and Business Media Deutschland GmbH, 2023. pp. 17-23 (Lecture Notes in Electrical Engineering).

@inproceedings{fcb3da7d6bb04d5298f6754e102b90ea,

title = "Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data",

abstract = "Due to the non-stationary randomness of the EEG signals, which results in a trained classifier with very low generalization ability. In order to improve the cross-session classification accuracy of Mental Workload (MW), a domain adaptive classification method is studied based on Transfer Component Analysis (TCA) using EEG Independent Components (ICs) in this paper. First step is to transform the EEG data in source and target domains in a same way. The TCA algorithm is used to build the transformation ϕ in order to minimize their Maximum Mean Discrepancy (MMD) distance between the source domain and the target domain. Second step is to train a MW classifier using the SVM and the transformed EEG data in the source domain. The target domain data are used for test. The presented method is compared with Subspace Alignment (SA). EEG data in the source and target domains are measured 24 h apart. A-distance is calculated to reveal the distribution difference between the transformed source and the target domains. The results show that the classification accuracy of the proposed method can be improved by 15.11\% compared with SA. The A-distance of TCA is significantly lower than that of SA.",

keywords = "Classification of mental workload, EEG, Independent Component Analysis (ICA), Subspace Alignment (SA), Transfer Component Analysis (TCA)",

author = "Hongquan Qu and Hanwen Dong and Liping Pang",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 22nd International Conference on Man-Machine-Environment System Engineering, MMESE 2022 ; Conference date: 21-10-2022 Through 23-10-2022",

year = "2023",

doi = "10.1007/978-981-19-4786-5\_3",

language = "英语",

isbn = "9789811947858",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "17--23",

editor = "Shengzhao Long and Dhillon, \{Balbir S.\}",

booktitle = "Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE",

address = "德国",

}

Qu, H, Dong, H & Pang, L 2023, Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data. in S Long & BS Dhillon (eds), Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE. Lecture Notes in Electrical Engineering, vol. 941 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 17-23, 22nd International Conference on Man-Machine-Environment System Engineering, MMESE 2022, Beijing, China, 21/10/22. https://doi.org/10.1007/978-981-19-4786-5_3

Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data. / Qu, Hongquan; Dong, Hanwen; Pang, Liping.
Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE. ed. / Shengzhao Long; Balbir S. Dhillon. Springer Science and Business Media Deutschland GmbH, 2023. p. 17-23 (Lecture Notes in Electrical Engineering; Vol. 941 LNEE).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data

AU - Qu, Hongquan

AU - Dong, Hanwen

AU - Pang, Liping

PY - 2023

Y1 - 2023

N2 - Due to the non-stationary randomness of the EEG signals, which results in a trained classifier with very low generalization ability. In order to improve the cross-session classification accuracy of Mental Workload (MW), a domain adaptive classification method is studied based on Transfer Component Analysis (TCA) using EEG Independent Components (ICs) in this paper. First step is to transform the EEG data in source and target domains in a same way. The TCA algorithm is used to build the transformation ϕ in order to minimize their Maximum Mean Discrepancy (MMD) distance between the source domain and the target domain. Second step is to train a MW classifier using the SVM and the transformed EEG data in the source domain. The target domain data are used for test. The presented method is compared with Subspace Alignment (SA). EEG data in the source and target domains are measured 24 h apart. A-distance is calculated to reveal the distribution difference between the transformed source and the target domains. The results show that the classification accuracy of the proposed method can be improved by 15.11% compared with SA. The A-distance of TCA is significantly lower than that of SA.

AB - Due to the non-stationary randomness of the EEG signals, which results in a trained classifier with very low generalization ability. In order to improve the cross-session classification accuracy of Mental Workload (MW), a domain adaptive classification method is studied based on Transfer Component Analysis (TCA) using EEG Independent Components (ICs) in this paper. First step is to transform the EEG data in source and target domains in a same way. The TCA algorithm is used to build the transformation ϕ in order to minimize their Maximum Mean Discrepancy (MMD) distance between the source domain and the target domain. Second step is to train a MW classifier using the SVM and the transformed EEG data in the source domain. The target domain data are used for test. The presented method is compared with Subspace Alignment (SA). EEG data in the source and target domains are measured 24 h apart. A-distance is calculated to reveal the distribution difference between the transformed source and the target domains. The results show that the classification accuracy of the proposed method can be improved by 15.11% compared with SA. The A-distance of TCA is significantly lower than that of SA.

KW - Classification of mental workload

KW - EEG

KW - Independent Component Analysis (ICA)

KW - Subspace Alignment (SA)

KW - Transfer Component Analysis (TCA)

UR - https://www.scopus.com/pages/publications/85136956822

U2 - 10.1007/978-981-19-4786-5_3

DO - 10.1007/978-981-19-4786-5_3

M3 - 会议稿件

AN - SCOPUS:85136956822

SN - 9789811947858

T3 - Lecture Notes in Electrical Engineering

SP - 17

EP - 23

BT - Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE

A2 - Long, Shengzhao

A2 - Dhillon, Balbir S.

PB - Springer Science and Business Media Deutschland GmbH

T2 - 22nd International Conference on Man-Machine-Environment System Engineering, MMESE 2022

Y2 - 21 October 2022 through 23 October 2022

ER -

Qu H, Dong H, Pang L. Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data. In Long S, Dhillon BS, editors, Man-Machine-Environment System Engineering - Proceedings of the 22nd International Conference on MMESE. Springer Science and Business Media Deutschland GmbH. 2023. p. 17-23. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-19-4786-5_3

Mental Workload Classification Method Based on Transfer Component Analysis with Cross-Session EEG Data

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this