EEG-based stress recognition using deep learning techniques

• Main Author: Ang, Jerica
• Other Authors: Wang Lipo
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/148896
• Institution: Nanyang Technological University

Being able to recognize early signs of mental stress is crucial towards preventing detrimental physical and/or mental effects on one’s health state. Electroencephalogram (EEG)-based stress recognition has been a commonly used method due to its many advantages over other physiological signals. However, there has yet to be an optimal deep learning technique for EEG-based stress recognition despite the many studies. This paper proposes using a popular supervised machine learning technique, Support Vector Machine (SVM) to detect stress. To provide a baseline for performance comparison, the results reported from another research paper with similar feature extraction method will be used. The highest classification accuracy obtained is 65.69%, detecting two levels of stress. Hopefully, this paper may be able to contribute to the ever-important research on detecting mental stress.