Social Information Improves Location Prediction in the Wild

How can knowing the location of my friends be used to more accurately predict my location? This paper ex-plores socially-aware location prediction under a partic-ularly challenging setting where the underlying interac-tions and social network are unknown and must be in-ferred over continuous spatiotemporal data. Our method samples inferred network topology using a linear re-gression model to predict future individual locations. We present an in-depth empirical study comparing dif-ferent network models and network sampling regimes under a bootstrapped sampling baseline. Furthermore, our qualitative analysis demonstrates the value of social information in population mobility modeling under our application’s challenges