Unobtrusive assessment of stress of office workers via analysis of their motion trajectories

E. Vildjiounaite (Corresponding Author), V. Huotari, Johanna Kallio, Vesa Kyllönen, Satu Marja Mäkelä, Georgy Gimel'farb

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Stress is the second most frequent work-related health problem in Europe, and it is important to assess stress risks relatively early to diminish negative consequences. Automatic detection of everyday stress is a challenging problem, however, as both stress perception and stress manifestation notably vary between individuals. Therefore stress detectors, trained on data of each person separately, usually achieve notably higher accuracies than non-personalised ones. Unfortunately, this accuracy gain requires collecting too large sets of labelled training data to realistically obtain from end users. In addition, the majority of current stress detectors exploit physiological or mobile phone data: the latter approach increases battery consumption compare with normal phone use, and the former requires to wear additional devices and to charge their batteries. Unlike previous work, this work proposes genuinely unobtrusive personalised stress detection system, based on use of environmental sensors and unsupervised training of hidden Markov models (HMM) classifier and hence requiring neither sensor maintenance nor data labelling efforts from end users. In the experiments with real life behavioural data of office workers, collected during 10 months, the proposed system achieved 67% accuracy of classifying each day as stressful vs. normal and 95% accuracy in classifying months, thanks to discovery of novel characteristics of motion trajectories, indicative of stress.

Original languageEnglish
Article number101028
JournalPervasive and Mobile Computing
Volume58
DOIs
Publication statusPublished - 1 Aug 2019
MoE publication typeA1 Journal article-refereed

Fingerprint

Trajectories
Trajectory
Motion
Battery
Detector
Detectors
Sensor
Sensors
Hidden Markov models
Medical problems
Mobile Phone
Mobile phones
Large Set
Labeling
Markov Model
Person
High Accuracy
Maintenance
Health
Classifiers

Cite this

@article{d47ad9e8573944039b4fd071d3a12f20,
title = "Unobtrusive assessment of stress of office workers via analysis of their motion trajectories",
abstract = "Stress is the second most frequent work-related health problem in Europe, and it is important to assess stress risks relatively early to diminish negative consequences. Automatic detection of everyday stress is a challenging problem, however, as both stress perception and stress manifestation notably vary between individuals. Therefore stress detectors, trained on data of each person separately, usually achieve notably higher accuracies than non-personalised ones. Unfortunately, this accuracy gain requires collecting too large sets of labelled training data to realistically obtain from end users. In addition, the majority of current stress detectors exploit physiological or mobile phone data: the latter approach increases battery consumption compare with normal phone use, and the former requires to wear additional devices and to charge their batteries. Unlike previous work, this work proposes genuinely unobtrusive personalised stress detection system, based on use of environmental sensors and unsupervised training of hidden Markov models (HMM) classifier and hence requiring neither sensor maintenance nor data labelling efforts from end users. In the experiments with real life behavioural data of office workers, collected during 10 months, the proposed system achieved 67{\%} accuracy of classifying each day as stressful vs. normal and 95{\%} accuracy in classifying months, thanks to discovery of novel characteristics of motion trajectories, indicative of stress.",
author = "E. Vildjiounaite and V. Huotari and Johanna Kallio and Vesa Kyll{\"o}nen and M{\"a}kel{\"a}, {Satu Marja} and Georgy Gimel'farb",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.pmcj.2019.05.009",
language = "English",
volume = "58",
journal = "Pervasive and Mobile Computing",
issn = "1574-1192",
publisher = "Elsevier",

}

Unobtrusive assessment of stress of office workers via analysis of their motion trajectories. / Vildjiounaite, E. (Corresponding Author); Huotari, V.; Kallio, Johanna; Kyllönen, Vesa; Mäkelä, Satu Marja; Gimel'farb, Georgy.

In: Pervasive and Mobile Computing, Vol. 58, 101028, 01.08.2019.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Unobtrusive assessment of stress of office workers via analysis of their motion trajectories

AU - Vildjiounaite, E.

AU - Huotari, V.

AU - Kallio, Johanna

AU - Kyllönen, Vesa

AU - Mäkelä, Satu Marja

AU - Gimel'farb, Georgy

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Stress is the second most frequent work-related health problem in Europe, and it is important to assess stress risks relatively early to diminish negative consequences. Automatic detection of everyday stress is a challenging problem, however, as both stress perception and stress manifestation notably vary between individuals. Therefore stress detectors, trained on data of each person separately, usually achieve notably higher accuracies than non-personalised ones. Unfortunately, this accuracy gain requires collecting too large sets of labelled training data to realistically obtain from end users. In addition, the majority of current stress detectors exploit physiological or mobile phone data: the latter approach increases battery consumption compare with normal phone use, and the former requires to wear additional devices and to charge their batteries. Unlike previous work, this work proposes genuinely unobtrusive personalised stress detection system, based on use of environmental sensors and unsupervised training of hidden Markov models (HMM) classifier and hence requiring neither sensor maintenance nor data labelling efforts from end users. In the experiments with real life behavioural data of office workers, collected during 10 months, the proposed system achieved 67% accuracy of classifying each day as stressful vs. normal and 95% accuracy in classifying months, thanks to discovery of novel characteristics of motion trajectories, indicative of stress.

AB - Stress is the second most frequent work-related health problem in Europe, and it is important to assess stress risks relatively early to diminish negative consequences. Automatic detection of everyday stress is a challenging problem, however, as both stress perception and stress manifestation notably vary between individuals. Therefore stress detectors, trained on data of each person separately, usually achieve notably higher accuracies than non-personalised ones. Unfortunately, this accuracy gain requires collecting too large sets of labelled training data to realistically obtain from end users. In addition, the majority of current stress detectors exploit physiological or mobile phone data: the latter approach increases battery consumption compare with normal phone use, and the former requires to wear additional devices and to charge their batteries. Unlike previous work, this work proposes genuinely unobtrusive personalised stress detection system, based on use of environmental sensors and unsupervised training of hidden Markov models (HMM) classifier and hence requiring neither sensor maintenance nor data labelling efforts from end users. In the experiments with real life behavioural data of office workers, collected during 10 months, the proposed system achieved 67% accuracy of classifying each day as stressful vs. normal and 95% accuracy in classifying months, thanks to discovery of novel characteristics of motion trajectories, indicative of stress.

UR - http://www.scopus.com/inward/record.url?scp=85068204359&partnerID=8YFLogxK

U2 - 10.1016/j.pmcj.2019.05.009

DO - 10.1016/j.pmcj.2019.05.009

M3 - Article

VL - 58

JO - Pervasive and Mobile Computing

JF - Pervasive and Mobile Computing

SN - 1574-1192

M1 - 101028

ER -