TY - GEN
T1 - Hierarchical span-based conditional random fields for labeling and segmenting events in wearable sensor data streams
AU - Adams, Roy J.
AU - Saleheen, Nazir
AU - Thomaz, Edison
AU - Parate, Abhinav
AU - Kumar, Santosh
AU - Marlin, Benjamin M.
N1 - Funding Information:
The authors would like to thank Deepak Ganesan and Gregory Abowd for helpful discussions and support of this research. This work was partially supported by the National Institutes of Health under awards R01DA033733, R01DA035502, 1R01CA190329, R01MD010362, and 1U54EB020404, and the National Science Foundation under awards IIS-1350522 and IIS-1231754.
PY - 2016
Y1 - 2016
N2 - The field of mobile health (mHealth) has the potential to yield new insights into health and behavior through the analysis of continuously recorded data from wearable health and activity sensors. In this paper, we present a hierarchical span-based conditional random field model for the key problem of jointly detecting discrete events in such sensor data streams and segmenting these events into high-level activity sessions. Our model includes higher-order cardinality factors and inter-event duration factors to capture domain-specific structure in the label space. We show that our model supports exact MAP inference in quadratic time via dynamic programming, which we leverage to perform learning in the structured support vector machine framework. We apply the model to the problems of smoking and eating detection using four real data sets. Our results show statistically significant improvements in segmentation performance relative to a hierarchical pairwise CRF.
AB - The field of mobile health (mHealth) has the potential to yield new insights into health and behavior through the analysis of continuously recorded data from wearable health and activity sensors. In this paper, we present a hierarchical span-based conditional random field model for the key problem of jointly detecting discrete events in such sensor data streams and segmenting these events into high-level activity sessions. Our model includes higher-order cardinality factors and inter-event duration factors to capture domain-specific structure in the label space. We show that our model supports exact MAP inference in quadratic time via dynamic programming, which we leverage to perform learning in the structured support vector machine framework. We apply the model to the problems of smoking and eating detection using four real data sets. Our results show statistically significant improvements in segmentation performance relative to a hierarchical pairwise CRF.
UR - http://www.scopus.com/inward/record.url?scp=84997637202&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84997637202&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84997637202
T3 - 33rd International Conference on Machine Learning, ICML 2016
SP - 529
EP - 538
BT - 33rd International Conference on Machine Learning, ICML 2016
A2 - Balcan, Maria Florina
A2 - Weinberger, Kilian Q.
PB - International Machine Learning Society (IMLS)
T2 - 33rd International Conference on Machine Learning, ICML 2016
Y2 - 19 June 2016 through 24 June 2016
ER -