Senior Data Science: Safe Aging with SPHERE Hosted By DrivenData


Woohoo! This competition has come to a close!

Many thanks to the participants for all of their hard work and commitment to using data for good!


This challenge is part of a large research project which centers around using sensors and algorithms to help older people live safely at home while maintaining their privacy and independence. Using passive, automated monitoring, the ultimate goal is to look out for a person's well-being without being burdensome or intrusive.

To gather data, researchers in the SPHERE Inter-disciplinary Research Collaboration (IRC) equipped volunteers with accelerometers similar to those found in cell phones or fitness wearables, and then had the subjects go about normal activities of daily living in a home-like environment that was also equipped with motion detectors. After gathering a robust set of sensor data, they had multiple annotators use camera footage to establish the ground truth, labeling chunks of sensor data as one of twenty specifically chosen activities (e.g. walk, sit, stand-to-bend, ascend stairs, descend stairs, etc).

Your challenge: help push forward the state of the art by predicting actual activity from sensor data!

Competition End Date:

July 31, 2016, 11:59 p.m. UTC

Place Prize Amount
1st $5,000
2nd $3,000
3rd $2,000


SPHERE is an interdiciplinary research colaboration between the Universities of Bristol, Reading, and Southampton. The vision of SPHERE is to impact healthcare needs simultaneously through data fusion and pattern recognition from a common platform of non-medical sensors in the home environment.

In this challenge, researchers are particularly interested in addressing the problem of robust sensor fusion, dealing with missing data and coping with inter-annotator disagreement for activity recognition. The dataset therefore consists of three sensing modalities (from depth cameras, accelerometers, and environmental sensors), and each sequence has been labeled by multiple annotators.


This challenge is taking place in conjunction with a discovery challenge workshop of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery. You don't have to be involved with the ECML/PKDD workshop in order to compete in this competition, but all workshop competitors will be entering their submissions here. Entrants should cite the following paper (reference given in BibTeX format):

@article{twomey2016sphere, title={The {SPHERE} Challenge: Activity Recognition with Multimodal Sensor Data}, author={Twomey, Niall and Diethe, Tom and Kull, Meelis and Song, Hao and Camplani, Massimo and Hannuna, Sion and Fafoutis, Xenofon and Zhu, Ni and Woznowski, Pete and Flach, Peter and Craddock, Ian}, journal={arXiv preprint arXiv:1603.00797}, year={2016}}

Winning entries will be offered the opportunity to submit papers to the discovery challenge. See this page for more details, and here for the workshop competition rules and eligibility.

AARP Foundation

This competition is supported by the great team at AARP Foundation. AARP Foundation works to ensure that low-income older adults have nutritious food, safe and affordable housing, a steady income, and strong and sustaining social bonds. We collaborate with individuals and organizations who share our commitment to innovation and our passion for problem-solving. Supported by vigorous legal advocacy, we create and advance effective solutions that help struggling older adults transform their lives. AARP Foundation is the charitable affiliate of AARP. Learn more at