Emotion Classification Based on EEG using Independent Component Analysis and Genetic Algorithm

Authors

  • Bimo Rian Tri Nugroho
  • Sugondo Hadiyoso
  • Rita Purnama Sari
  • Inung Wijayanto

Abstract

Emotion is a psychological condition as a reaction to an event. All forms of emotions are thought to be controlled by the central nervous system, the brain. A human’s emotions are often associated with changes in facial expressions as markers. Several studies have successfully carried out emotional detection through facial imaging. But the detection of emotions through these expressions tends to be hidden. Therefore, in this research study emotions through EEG signals to overcome these problems because EEG characteristics are unique and difficult to hide. Emotional EEG signals come from the DEAP database in the form of arousal, valence, liking, dominance, and familiarity with video as the stimulus. The raw EEG signal processed by Independent Component Analysis (ICA) to obtain features is then classified by Support Vector Machine (SVM) combined with Genetic Algorithm (GA) for optimization. The simulation carried out gave an accuracy of 77.27% on the classification of sad and happy emotions.

Downloads

Published

2020-04-09

Issue

Section

Articles