Feature Comparison of Emotion Estimation by EEG and Heart Rate Variability Indices and Accuracy Evaluation by Machine Learning

研究成果: Conference contribution

抄録

There has been a lot of attempts on estimating human emotions using physio-logical data, and it is expected to be applied to medical diagnosis. Recently, there is emotion estimation model using EEG and heart rate variability index-es as feature values, and applying deep learning to classify emotions with an accuracy of 61%. However, the accuracy may not be sufficient for applications such as medical diagnosis. In this study, we extracted and selected features of EEG and heart rate variability indexes in order to improve the accuracy. As a result, by using our proposed method to extract and select features, the accuracy of the model was increased to almost 100%.

本文言語English
ホスト出版物のタイトルAdvances in Neuroergonomics and Cognitive Engineering - Proceedings of the AHFE 2021 Virtual Conferences on Neuroergonomics and Cognitive Engineering, Industrial Cognitive Ergonomics and Engineering Psychology, and Cognitive Computing and Internet of Things, 2021
編集者Hasan Ayaz, Umer Asgher, Lucas Paletta
出版社Springer Science and Business Media Deutschland GmbH
ページ222-230
ページ数9
ISBN(印刷版)9783030802844
DOI
出版ステータスPublished - 2021
イベントAHFE Conferences on Neuroergonomics and Cognitive Engineering, Industrial Cognitive Ergonomics and Engineering Psychology, and Cognitive Computing and Internet of Things, 2021 - Virtual, Online
継続期間: 2021 7月 252021 7月 29

出版物シリーズ

名前Lecture Notes in Networks and Systems
259
ISSN(印刷版)2367-3370
ISSN(電子版)2367-3389

Conference

ConferenceAHFE Conferences on Neuroergonomics and Cognitive Engineering, Industrial Cognitive Ergonomics and Engineering Psychology, and Cognitive Computing and Internet of Things, 2021
CityVirtual, Online
Period21/7/2521/7/29

ASJC Scopus subject areas

  • 制御およびシステム工学
  • 信号処理
  • コンピュータ ネットワークおよび通信

フィンガープリント

「Feature Comparison of Emotion Estimation by EEG and Heart Rate Variability Indices and Accuracy Evaluation by Machine Learning」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル