Change search
ReferencesLink to record
Permanent link

Direct link
Wearable Sensor Data Fusion for Human Stress Estimation
Linköping University, Department of Electrical Engineering, Automatic Control.
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
Fusion av data från bärbara sensorer för estimering av mänsklig stress (Swedish)
Abstract [en]

With the purpose of classifying and modelling stress, different sensors, signal features, machine learning methods, and stress experiments have been compared. Two databases have been studied: the MIT driver stress database and a new experimental database, where three stress tasks have been performed for 9 subjects: the Trier Social Stress Test, the Socially Evaluated Cold Pressor Test and the d2 test, of which the latter is not classically used for generating stress. Support vector machine, naive Bayes, k-nearest neighbor and probabilistic neural network classification techniques were compared, with support vector machines achieving the highest performance in general (99.5 ±0.6 %$on the driver database and 91.4 ± 2.4 % on the experimental database). For both databases, relevant features include the mean of the heart rate and the mean of the galvanic skin response, together with the mean of the absolute derivative of the galvanic skin response signal. A new feature is also introduced with great performance in stress classification for the driver database. Continuous models for estimating stress levels have also been developed, based upon the perceived stress levels given by the subjects during the experiments, where support vector regression is more accurate than linear and variational Bayesian regression.

Abstract [sv]

I syfte att klassificera och modellera stress har olika sensorer, signalegenskaper, maskininlärningsmetoder och stressexperiment jämförts. Två databaser har studerats: MIT:s förarstressdatabas och en ny databas baserad på egna experiment, där stressuppgifter har genomförts av nio försökspersoner: Trier Social Stress Test,  Socially Evaluated Cold Pressor Test och d2-testet, av vilka det sistnämnda inte normalt används för att generera stress. Support vector machine-, naive Bayes-, k-nearest neighbour- och probabilistic neural network-algoritmer har jämförts, av vilka support vector machine har uppnått den högsta prestandan i allmänhet (99.5 ± 0.6 % på förardatabasen, 91.4 ± 2.4 %  på experimenten). För båda databaserna har signalegenskaper såsom medelvärdet av hjärtrytmen och hudens ledningsförmåga, tillsammans med medelvärdet av beloppet av hudens ledningsförmågas derivata identifierats som relevanta. En ny signalegenskap har också introducerats, med hög prestanda i stressklassificering på förarstressdatabasen. En kontinuerlig modell har också utvecklats, baserad på den upplevda stressnivån angiven av försökspersonerna under experimenten, där support vector regression har uppnått bättre resultat än linjär regression och variational Bayesian regression.

Place, publisher, year, edition, pages
2015. , 123 p.
Keyword [en]
stress, data fusion, classification, modelling, wearables, physiological sensors
National Category
Control Engineering
URN: urn:nbn:se:liu:diva-122348ISRN: LiTH-ISY-EX–15/4904–SEOAI: diva2:865706
External cooperation
Commissariat à l'Énergie Atomique et aux Énergies Alternatives
Subject / course
Automatic Control
2015-10-26, Systemet, ISY, Linköping University, SE-581 83, LINKÖPING, 11:05 (English)
Available from: 2015-10-30 Created: 2015-10-29 Last updated: 2015-10-30Bibliographically approved

Open Access in DiVA

Wearable Sensor Data Fusion for Human Stress Estimation(16401 kB)733 downloads
File information
File name FULLTEXT01.pdfFile size 16401 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Ollander, Simon
By organisation
Automatic Control
Control Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 733 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 1928 hits
ReferencesLink to record
Permanent link

Direct link