Analyzing group interaction on networks of face-to-face proximity using wearable sensors

The Internet of Things as well as smart devices enable the capture of multi-modal social interaction data at large-scale. For understanding the behavior of the involved actors, as well as to enable structural modeling, the analysis of the according social interaction networks is essential. In contrast to standard approaches that capture social network data using questionaires, this paper describes an analysis at larger scale using sensor data collected by Radio Frequency Identification (RFID) tags. We complement it by informant self-report data obtained using surveys. We focus on the social network of a students' freshman week, and investigate research questions concerning the communication behavior and structure, gender homophily, and inter-relations of sensor-based (RFID) and self-report social networks.