University of Minnesota
Passive Stress Detection and Prediction in Naturalistic Environments Using Wearable Sensors and Machine Learning
Abstract
dc:description.abstractStress is an important problem in modern society. Stress detection has been investigated in laboratory settings, however it is still challenging to detect or predict stress events in naturalistic environments due to the complexity introduced by daily life. However, widely available commercial wearable sensors enable collection of high quality physiological data in naturalistic environments. In this study, I utilized physiological data collected from wearable sensors to detect patterns during and before stress events. Stress events are associated with physiological changes and also environmental factors. Features were calculated from physiological signals such as electrocardiogram, photoplethysmography, electrodermal activity and accelerometer and GPS data. A wide range of ML algorithms were applied to these features for investigating the feasibility of detecting and predicting stress and stress related behavior.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yu, Hang
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Yu, Hang. (2025). Passive Stress Detection and Prediction in Naturalistic Environments Using Wearable Sensors and Machine Learning. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/279142.