1. Home
  2. Proceedings
  3. BIBM
  4. BIBM 2022
2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

Prevent Over-fitting and Redundancy in Physiological Signal Analyses for Stress Detection

Year: 2022, Pages: 2585-2588

Authors

Ruijie Fang, University of California,ECE Department,Davis,CA,USA  
Ruoyu Zhang, University of California,ECE Department,Davis,CA,USA  
Elahe Hosseini, University of California,ECE Department,Davis,CA,USA  
Anna M. Parenteau, University of California,Department of Psychology,Davis,CA,USA  
Sally Hang, University of California,Department of Psychology,Davis,CA,USA  
Setareh Rafatirad, University of California,ECE Department,Davis,CA,USA  
Camelia E. Hostinar, University of California,Department of Psychology,Davis,CA,USA  
Mahdi Orooji, University of California,ECE Department,Davis,CA,USA  
Houman Homayoun, University of California,ECE Department,Davis,CA,USA
Download PDF

Abstract

Stress detection is an emerging field. WESAD is a commonly used public dataset for automated stress detection. It contains physiological signals including ECG, EDA, EMG, ACC, BVP, EDA, and skin temperature. The time window approach is used to extract features from time-series physiological signals. We find in previous studies that a 60-second time window with a 0.25-second window shift is widely used, but such window settings may cause redundancy and over-fitting. Thus, we propose to use (1) new window settings and (2) normalization per subject to tackle this problem. The experiment results show that our proposed methods significantly increase the classification performance.
Like what you’re reading?
Already a member?
Get this article FREE with a new membership!

Related Articles