aalto1 untyped-item.component.html
Emotion recognition based on physiological signal
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
Department
Major/Subject
Mcode
ELEC3029
Language
en
Pages
59+4
Series
Abstract
This research focuses on emotion recognition based on physiological signals, specifically electroencephalogram (EEG) signals and heart rate data. Utilizing virtual reality (VR) technology to evoke genuine emotional responses, the study collected data from 14 subjects across different countries. The primary objective was to evaluate emotion classification algorithms, analyzing their accuracy in various scenarios. Employing machine learning and deep learning, including support vector machines (SVM) and deep learning models, the results demonstrated significant accuracy, reaching up to 88% in single dimensions (arousal or valence) and 84% in dual dimensions (arousal and valence together as labels). Multi-feature combinations, especially in deep learning models, enhanced sentiment classification. Identified areas for improvement include the quality of VR stimuli, additional physiological signals, and refined experimental procedures. Future research will explore algorithm stacking and integration methods to further enhance emotion recognition accuracy.