A survey of location-based social networks: problems, methods, and future research directions

The development of mobile devices and positioning technology has facilitated the rapid growth of location-based social networks (LBSNs). Users in these networks can share geo-related information in real-time, including locations, trajectories, geo-tagged pictures, and tweets. LBSNs record massive amounts of spatiotemporal data and offer a great opportunity to analyze human and location-specific spatiotemporal characteristics. It plays an important role in various applications, such as marketing, recommendations, and urban planning. In this study, we collect relevant literature about LBSNs research in the past 10 years and use a topic model, latent Dirichlet allocation (LDA), to uncover the highly heterogeneous area of research related to LBSNs. Then, we conduct a systematic review of those works. In doing so, we organize identified literature into eight fine-grained directions. For each direction, we sum up the major research focus and contributions. We also systematize future research into four main themes concerning data simulation and fusion, privacy-aware methods, new applications and services, and technological innovations.