TY - GEN
T1 - An asynchronous multi-view learning approach for activity recognition using wearables
AU - Ma, Yuchao
AU - Ghasemzadeh, Hassan
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/10/13
Y1 - 2016/10/13
N2 - In this paper, we introduce an Asynchronous Multiview Learning (AML) approach to allow accurate transfer of activity classification models across asynchronous sensor views. Our study is motivated by the highly dynamic nature of health monitoring using wearable sensors. Such dynamics include changes in sensing platform (e.g., sensor upgrade) and platform settings (e.g., sampling frequency, on-body sensor location), which result in failure of the machine learning algorithms if they remain untrained in the new setting. Our approach allows machine learning algorithms to automatically reconfigure without any need for labeled training data in the new setting. Our evaluation using real data collected with wearable motion sensors demonstrates that the average classification accuracy using our automatically labeled training data is 85.2%. This accuracy is only 3.4% to 4.5% less than the experimental upper bound, where ground truth labeled training data are used to develop a new activity recognition classifier.
AB - In this paper, we introduce an Asynchronous Multiview Learning (AML) approach to allow accurate transfer of activity classification models across asynchronous sensor views. Our study is motivated by the highly dynamic nature of health monitoring using wearable sensors. Such dynamics include changes in sensing platform (e.g., sensor upgrade) and platform settings (e.g., sampling frequency, on-body sensor location), which result in failure of the machine learning algorithms if they remain untrained in the new setting. Our approach allows machine learning algorithms to automatically reconfigure without any need for labeled training data in the new setting. Our evaluation using real data collected with wearable motion sensors demonstrates that the average classification accuracy using our automatically labeled training data is 85.2%. This accuracy is only 3.4% to 4.5% less than the experimental upper bound, where ground truth labeled training data are used to develop a new activity recognition classifier.
UR - http://www.scopus.com/inward/record.url?scp=85009075233&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85009075233&partnerID=8YFLogxK
U2 - 10.1109/EMBC.2016.7591386
DO - 10.1109/EMBC.2016.7591386
M3 - Conference contribution
C2 - 28268968
AN - SCOPUS:85009075233
T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
SP - 3105
EP - 3108
BT - 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016
Y2 - 16 August 2016 through 20 August 2016
ER -