Spatio-temporal patterns link your digital identities

摘要: An important challenge for mobility analysis is the development of techniques that can associate users’ identities across multiple datasets. These assist in developing hybrid sensing and tracking mechanisms large urban spaces, inferring context by combining datasets, but at same time have implications privacy. In this paper we present a scheme to different person two movement databases. Our key contributions are reformulation problem terms two-class classification, efficient pruning search space. We evaluate performance on synthetic real data from co-located city-wide WiFi Bluetooth networks, show has remarkable effect identifying individuals distinct Finally, discuss privacy light our findings.