Home // COGNITIVE 2014, The Sixth International Conference on Advanced Cognitive Technologies and Applications // View article

Emotion Classification Based on Bio-Signals Using Machine Learning Algorithms

Authors:
Eun-Hye Jang
Byoung-Jun Park
Sang-Hyeob Kim
Myung-Ae Chung
Yeongji Eum
Jin-Hun Sohn

Keywords: bio-signal; feature extraction; machine learning algorithm

Abstract:
In human-computer interaction researches, one of the most interesting topics in the field of emotion recognition is to recognize human's feeling using bio-signals. According to previous researches, it is known that there is strong correlation between human emotion state and physiological reaction. Bio-signals takes noticed lately because those can be simply acquired with some sensors and are less sensitive in social and cultural difference. We have applied four algorithms, linear discriminant analysis, Naïve Bayes, decision tree and support vector machine to classify emotions, happiness, anger, surprise and stress based on bio-signals. In this study, audio-visual film clips were used to evoke each emotion and bio-signals (electrocardiograph, electrodermal activity, photoplethysmo-graph, and skin temperature) as emotional responses were measured and the features were extracted from them. For emotion recognition, the used algorithms are evaluated by only training, 10-fold cross-validation and repeated random sub-sampling validation. We have obtained very low recognition accuracy from 28.0 to 38.4% for testing. This means that it needs to apply various methodologies for the accuracy improvement of emotion recognition in the future analysis. Nevertherless, this can be helpful to provide the basis for the emotion recognition technique in human-machine interaction as well as contribute to the standardization in emotion-specific autonomic nervous system responses.

Pages: 104 to 109

Copyright: Copyright (c) IARIA, 2014

Publication date: May 25, 2014

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-340-7

Location: Venice, Italy

Dates: from May 25, 2014 to May 29, 2014