The back-propagation neural network classification of EEG signal using time frequency domain feature extraction

Diah K. Theresia, Dessy Ana, Akhmad Faqih, Benyamin Kusumoputro

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the recognition of emotions based on EEG signals, the feature extraction used is as important as classifier for the results of classification. The better the feature extraction is used, the better the results of the classification. In the other hand, there is no definitive approach for feature extraction in emotion recognition based on EEG signal. In this paper, we use nine types of time frequency domains as features, Principal Component Analysis (PCA) as dimension reduction method and Back-propagation Neural Network as classifiers. This method is implemented using a database that can be accessed by the public. The results of the experiment show that time frequency domain feature extraction and back propagation can achieve 63.75 % recognition rate.

Original languageEnglish
Title of host publication2019 16th International Conference on Quality in Research, QIR 2019 - International Symposium on Electrical and Computer Engineering
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728118987
DOIs
Publication statusPublished - Jul 2019
Event16th International Conference on Quality in Research, QIR 2019 - Padang, Indonesia
Duration: 22 Jul 201924 Jul 2019

Publication series

Name2019 16th International Conference on Quality in Research, QIR 2019 - International Symposium on Electrical and Computer Engineering

Conference

Conference16th International Conference on Quality in Research, QIR 2019
CountryIndonesia
CityPadang
Period22/07/1924/07/19

Keywords

  • Back-propagation Neural Network
  • DEAP
  • EEG
  • PCA
  • Time Variant Feature

Fingerprint Dive into the research topics of 'The back-propagation neural network classification of EEG signal using time frequency domain feature extraction'. Together they form a unique fingerprint.

Cite this