Physical activity recognition from accelerometer data using a multi-scale ensemble method

Zheng, Yonglei, Wong, Weng-Keen, Guan, Xinze and Trost, Stewart (2013). Physical activity recognition from accelerometer data using a multi-scale ensemble method. In: Hector Muñoz-Avila and David J. Stracuzzi, Proceedings of the 25th Annual Conference on Innovative Applications of Artificial Intelligence. IAAI-13: The Twenty-Fifth Annual Conference on Innovative Applications of Artificial Intelligence, Bellevue, WA, United States, (1575-1581). 14–18 July 2013.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Author Zheng, Yonglei
Wong, Weng-Keen
Guan, Xinze
Trost, Stewart
Title of paper Physical activity recognition from accelerometer data using a multi-scale ensemble method
Conference name IAAI-13: The Twenty-Fifth Annual Conference on Innovative Applications of Artificial Intelligence
Conference location Bellevue, WA, United States
Conference dates 14–18 July 2013
Proceedings title Proceedings of the 25th Annual Conference on Innovative Applications of Artificial Intelligence
Place of Publication Palo Alto, CA, United States
Publisher Association for the Advancement of Artificial Intelligence (AAAI)
Publication Year 2013
Sub-type Fully published paper
Open Access Status
ISBN 9781577356158
Editor Hector Muñoz-Avila
David J. Stracuzzi
Start page 1575
End page 1581
Total pages 7
Collection year 2014
Language eng
Formatted Abstract/Summary
Accurate and detailed measurement of an individual's physical activity is a key requirement for helping researchers understand the relationship between physical activity and health. Accelerometers have become the method of choice for measuring physical activity due to their small size, low cost, convenience and their ability to provide objective information about physical activity. However, interpreting accelerometer data once it has been collected can be challenging. In this work, we applied machine learning algorithms to the task of physical activity recognition from triaxial accelerometer data. We employed a simple but effective approach of dividing the accelerometer data into short non-overlapping windows, converting each window into a feature vector, and treating each feature vector as an i.i.d training instance for a supervised learning algorithm. In addition, we improved on this simple approach with a multi-scale ensemble method that did not need to commit to a single window size and was able to leverage the fact that physical activities produced time series with repetitive patterns and discriminative features for physical activity occurred at different temporal scales.
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Sun, 12 Jan 2014, 16:11:33 EST by Deborah Noon on behalf of School of Human Movement and Nutrition Sciences