Classification of Working Memory Performance from EEG with Deep Artificial Neural Networks

Youngchul Kwak, Woo Jin Song, Seong Eun Kim

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

3 Scopus citations

Abstract

Individuals have different working memory performance and some studies investigated a relationship between working memory performance and electroencephalography (EEG) band power. In this paper, we study EEG features to classify low performance group and high performance group and find that the power ratio feature of alpha and beta is more separable than their absolute powers. We test a deep artificial neural network (ANN) using the power ratio feature to classify the low performance group and high performance group. Experimental results on the working memory tasks show that some subjects have quite low accuracies (<20%) and it results in a low average classification accuracy of 61%, but we can see a possibility in the estimation of working memory performance using EEG data.

Original languageEnglish
Title of host publication7th International Winter Conference on Brain-Computer Interface, BCI 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538681169
DOIs
StatePublished - Feb 2019
Event7th International Winter Conference on Brain-Computer Interface, BCI 2019 - Gangwon, Korea, Republic of
Duration: 18 Feb 201920 Feb 2019

Publication series

Name7th International Winter Conference on Brain-Computer Interface, BCI 2019

Conference

Conference7th International Winter Conference on Brain-Computer Interface, BCI 2019
Country/TerritoryKorea, Republic of
CityGangwon
Period18/02/1920/02/19

Keywords

  • Artificial neural network (ANN)
  • EEG band power
  • working memory

Fingerprint

Dive into the research topics of 'Classification of Working Memory Performance from EEG with Deep Artificial Neural Networks'. Together they form a unique fingerprint.

Cite this