Segmenting accelerometer data from daily life with unsupervised machine learning.

Journal: PloS one
Published Date: Jan 9, 2019

Abstract

PURPOSE: Accelerometers are increasingly used to obtain valuable descriptors of physical activity for health research. The cut-points approach to segment accelerometer data is widely used in physical activity research but requires resource expensive calibration studies and does not make it easy to explore the information that can be gained for a variety of raw data metrics. To address these limitations, we present a data-driven approach for segmenting and clustering the accelerometer data using unsupervised machine learning.

Authors

  • Dafne van Kuppevelt
    Netherlands eScience Center, Amsterdam, The Netherlands.
  • Joe Heywood
    Centre for Longitudinal Studies, UCL Institute of Education, London, United Kingdom.
  • Mark Hamer
    School Sport, Exercise and Health Sciences, Loughborough University, Loughborough, United Kingdom.
  • Séverine Sabia
    INSERM, U1018, Centre for Research in Epidemiology and Population Health, Paris, France.
  • Emla Fitzsimons
    Centre for Longitudinal Studies, UCL Institute of Education, London, United Kingdom.
  • Vincent van Hees
    Netherlands eScience Center, Amsterdam, The Netherlands.

Keywords

External Resources

Popular Topics
Recent Journals