Wearable health care devices, gadgets for monitoring personal activity including steps to heart rates, are gaining popularity on health-care and sport fields. Consequently, precise analysis of the health and physical performance of the individual is promising.
In contrast, our research focuses on expanding the data analysis from “individual(s)” to “groups” and “communities”.
As a first step of the data analysis for “group” level analysis, we developed a method to detect people moving around together by using only the step tracking data acquired from smart pedometers like Fitbit. From the step data collected in our evaluation, we confirmed that working-relationships can be visualized by detecting people moving around.
Kota Tsubouchi, Ryoma Kawajiri, Masamichi Shimosaka
Working-relationship detection from fitbit sensor data
UbiComp ’13 Adjunct Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publication, 2013
Fine-grained Social Relationship Extraction from Real Activity Data under Coarse Supervision
In proceedings of the 2015 ACM International Symposium on Wearable Computers, 2015.
Kota Tsubouchi (Yahoo! JAPAN Corporation), Ryoma Kawajiri, Masamichi Shimosaka.