Abstract:
This systematic review groups and summarizes the algorithms used to detect the state of stress and the physiological and behavioral features used to feed these algorithms...View moreMetadata
Abstract:
This systematic review groups and summarizes the algorithms used to detect the state of stress and the physiological and behavioral features used to feed these algorithms, associated with each type of stressor. Method: We conducted a systematic literature review following the PRISMA-statement in seven databases. The studies considered for this review were those that used physiological or behavioral response to detect stress state. Results: 27 publications (29 independent studies) were included in the review. Stress detection accuracy ranged from 54% using a Decision Tree (DT) to 100% using Linear Discriminant Analysis (LDA). 72.4% studies used psychological stressors and 27.6% physical stressors to generate the stress state. Conclusions: The behavioral response has not been widely studied in stress detection and may be the key to identify which stressor is generating a particular stress state.
Published in: 2021 9th International Conference on Affective Computing and Intelligent Interaction (ACII)
Date of Conference: 28 September 2021 - 01 October 2021
Date Added to IEEE Xplore: 15 November 2021
ISBN Information: