Stable feature selection for EEG-based emotion recognition

Zirui Lan, Olga Sourina, Lipo Wang, Yisi Liu, Reinhold Scherer, Gernot R. Müller-Putz

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Affective brain-computer interface (aBCI) introduces personal affective factors into human-computer interactions, which could potentially enrich the user's experience during the interaction with a computer. However, affective neural patterns are volatile even within the same subject. To maintain satisfactory emotion recognition accuracy, the state-of-the-art aBCIs mainly tailor the classifier to the subject-of-interest and require frequent re-calibrations for the classifier. In this paper, we demonstrate that the recognition accuracy of aBCIs deteriorates when re-calibration is ruled out during the long-term usage for the same subject. Then, we propose a stable feature selection method to choose the most stable affective features, for mitigating the accuracy deterioration to a lesser extent and maximizing the aBCI performance in the long run. We validate our method on a dataset comprising six subjects' EEG data collected during two sessions per day for each subject for eight consecutive days.

Original languageEnglish
Title of host publicationProceedings - 2018 International Conference on Cyberworlds, CW 2018
EditorsAlexei Sourin, Olga Sourina, Marius Erdt, Christophe Rosenberger
PublisherInstitute of Electrical and Electronics Engineers
Pages176-183
Number of pages8
ISBN (Electronic)9781538673157
DOIs
Publication statusPublished - 26 Dec 2018
Event17th International Conference on Cyberworlds, CW 2018 - Singapore, Singapore
Duration: 3 Oct 20185 Oct 2018

Conference

Conference17th International Conference on Cyberworlds, CW 2018
CountrySingapore
CitySingapore
Period3/10/185/10/18

Fingerprint

Emotion Recognition
Brain computer interface
Electroencephalography
Feature Selection
Feature extraction
Classifiers
Calibration
Human computer interaction
Deterioration
Classifier
Volatiles
User Experience
Long-run
Interaction
Consecutive
Choose
Electroencephalogram
Demonstrate
Brain

Keywords

  • Electroencephalography (EEG)
  • Emotion recognition
  • Feature selection
  • Intra correlation coefficient (ICC)
  • Stable feature

ASJC Scopus subject areas

  • Signal Processing
  • Modelling and Simulation
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fields of Expertise

  • Human- & Biotechnology

Cite this

Lan, Z., Sourina, O., Wang, L., Liu, Y., Scherer, R., & Müller-Putz, G. R. (2018). Stable feature selection for EEG-based emotion recognition. In A. Sourin, O. Sourina, M. Erdt, & C. Rosenberger (Eds.), Proceedings - 2018 International Conference on Cyberworlds, CW 2018 (pp. 176-183). [8590037] Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CW.2018.00042

Stable feature selection for EEG-based emotion recognition. / Lan, Zirui; Sourina, Olga; Wang, Lipo; Liu, Yisi; Scherer, Reinhold; Müller-Putz, Gernot R.

Proceedings - 2018 International Conference on Cyberworlds, CW 2018. ed. / Alexei Sourin; Olga Sourina; Marius Erdt; Christophe Rosenberger. Institute of Electrical and Electronics Engineers, 2018. p. 176-183 8590037.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Lan, Z, Sourina, O, Wang, L, Liu, Y, Scherer, R & Müller-Putz, GR 2018, Stable feature selection for EEG-based emotion recognition. in A Sourin, O Sourina, M Erdt & C Rosenberger (eds), Proceedings - 2018 International Conference on Cyberworlds, CW 2018., 8590037, Institute of Electrical and Electronics Engineers, pp. 176-183, 17th International Conference on Cyberworlds, CW 2018, Singapore, Singapore, 3/10/18. https://doi.org/10.1109/CW.2018.00042
Lan Z, Sourina O, Wang L, Liu Y, Scherer R, Müller-Putz GR. Stable feature selection for EEG-based emotion recognition. In Sourin A, Sourina O, Erdt M, Rosenberger C, editors, Proceedings - 2018 International Conference on Cyberworlds, CW 2018. Institute of Electrical and Electronics Engineers. 2018. p. 176-183. 8590037 https://doi.org/10.1109/CW.2018.00042
Lan, Zirui ; Sourina, Olga ; Wang, Lipo ; Liu, Yisi ; Scherer, Reinhold ; Müller-Putz, Gernot R. / Stable feature selection for EEG-based emotion recognition. Proceedings - 2018 International Conference on Cyberworlds, CW 2018. editor / Alexei Sourin ; Olga Sourina ; Marius Erdt ; Christophe Rosenberger. Institute of Electrical and Electronics Engineers, 2018. pp. 176-183
@inproceedings{50c46cc99283481b906bcb8320b3733f,
title = "Stable feature selection for EEG-based emotion recognition",
abstract = "Affective brain-computer interface (aBCI) introduces personal affective factors into human-computer interactions, which could potentially enrich the user's experience during the interaction with a computer. However, affective neural patterns are volatile even within the same subject. To maintain satisfactory emotion recognition accuracy, the state-of-the-art aBCIs mainly tailor the classifier to the subject-of-interest and require frequent re-calibrations for the classifier. In this paper, we demonstrate that the recognition accuracy of aBCIs deteriorates when re-calibration is ruled out during the long-term usage for the same subject. Then, we propose a stable feature selection method to choose the most stable affective features, for mitigating the accuracy deterioration to a lesser extent and maximizing the aBCI performance in the long run. We validate our method on a dataset comprising six subjects' EEG data collected during two sessions per day for each subject for eight consecutive days.",
keywords = "Electroencephalography (EEG), Emotion recognition, Feature selection, Intra correlation coefficient (ICC), Stable feature",
author = "Zirui Lan and Olga Sourina and Lipo Wang and Yisi Liu and Reinhold Scherer and M{\"u}ller-Putz, {Gernot R.}",
year = "2018",
month = "12",
day = "26",
doi = "10.1109/CW.2018.00042",
language = "English",
pages = "176--183",
editor = "Alexei Sourin and Olga Sourina and Marius Erdt and Christophe Rosenberger",
booktitle = "Proceedings - 2018 International Conference on Cyberworlds, CW 2018",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

TY - GEN

T1 - Stable feature selection for EEG-based emotion recognition

AU - Lan, Zirui

AU - Sourina, Olga

AU - Wang, Lipo

AU - Liu, Yisi

AU - Scherer, Reinhold

AU - Müller-Putz, Gernot R.

PY - 2018/12/26

Y1 - 2018/12/26

N2 - Affective brain-computer interface (aBCI) introduces personal affective factors into human-computer interactions, which could potentially enrich the user's experience during the interaction with a computer. However, affective neural patterns are volatile even within the same subject. To maintain satisfactory emotion recognition accuracy, the state-of-the-art aBCIs mainly tailor the classifier to the subject-of-interest and require frequent re-calibrations for the classifier. In this paper, we demonstrate that the recognition accuracy of aBCIs deteriorates when re-calibration is ruled out during the long-term usage for the same subject. Then, we propose a stable feature selection method to choose the most stable affective features, for mitigating the accuracy deterioration to a lesser extent and maximizing the aBCI performance in the long run. We validate our method on a dataset comprising six subjects' EEG data collected during two sessions per day for each subject for eight consecutive days.

AB - Affective brain-computer interface (aBCI) introduces personal affective factors into human-computer interactions, which could potentially enrich the user's experience during the interaction with a computer. However, affective neural patterns are volatile even within the same subject. To maintain satisfactory emotion recognition accuracy, the state-of-the-art aBCIs mainly tailor the classifier to the subject-of-interest and require frequent re-calibrations for the classifier. In this paper, we demonstrate that the recognition accuracy of aBCIs deteriorates when re-calibration is ruled out during the long-term usage for the same subject. Then, we propose a stable feature selection method to choose the most stable affective features, for mitigating the accuracy deterioration to a lesser extent and maximizing the aBCI performance in the long run. We validate our method on a dataset comprising six subjects' EEG data collected during two sessions per day for each subject for eight consecutive days.

KW - Electroencephalography (EEG)

KW - Emotion recognition

KW - Feature selection

KW - Intra correlation coefficient (ICC)

KW - Stable feature

UR - http://www.scopus.com/inward/record.url?scp=85061436441&partnerID=8YFLogxK

U2 - 10.1109/CW.2018.00042

DO - 10.1109/CW.2018.00042

M3 - Conference contribution

SP - 176

EP - 183

BT - Proceedings - 2018 International Conference on Cyberworlds, CW 2018

A2 - Sourin, Alexei

A2 - Sourina, Olga

A2 - Erdt, Marius

A2 - Rosenberger, Christophe

PB - Institute of Electrical and Electronics Engineers

ER -