Inferring Real-World Relationships from Spatiotemporal Data

The pervasiveness of GPS-enabled mobile devices and the popularity of location-based services have generated, for the first time, massive data that represents the movements of people in the real world at a high resolution, aka spatiotemporal data. Such collections of spatiotemporal data constitute a rich source of information for studying various social behaviors, and in particular, give a boost to the study of inferring the real-world social connections from spatiotemporal data. This article surveys the prominent techniques proposed for deriving social connections and social strength from spatiotemporal data and discusses their formulations.