Leveraging Bluetooth co-location traces in group discovery algorithms

作者：Boston Daniel^*; Mardenfeld Steve; Pan Juan; Jones Quentin; Iamnitchi Adriana; Borcea Cristian

来源：Pervasive and Mobile Computing, 2014, 11: 88-105.

DOI：10.1016/j.pmcj.2012.10.003

• Mobile social computing
• Group discovery
• Co-location traces
• Smart phones
• Distributed computing

Smart phones can collect and share Bluetooth co-location traces to identify ad hoc or semipermanent social groups, which can enhance recommender systems or allow detection of epidemic events. Group discovery using Bluetooth co-location is practical due to low power consumption, short range, and applicability to decentralization. This paper presents the Group Discovery using Co-location traces (GDC) and Decentralized GDC (DGDC) algorithms, which leverage user meeting frequency and duration to accurately detect groups. GDC and DGDC are validated on one month of data collected from 141 smart phones carried by students on our campus, and by comparison against ground truth groups.

• 出版日期2014-4