Community detection in real communication using body sway and voice information with ambient sensing chair

The authors aim to develop a community detection system without constraining users. This paper introduces a method to detect communities in which users are engaging in a conversation with each other, by an ambient sensing of body sways and their synchrony between users. We apply an improved ambient sensing chair, which had been developed in our previous study. The system employs wireless force sensors to measure the center of pressure. By using measured speaker/listener information and synchrony between any users´ body sways, it is confirmed to detect the community information in experiment.