Privacy and the City: User Identification and Location Semantics in Location-Based Social Networks

With the advent of GPS enabled smartphones, an increasing number of users is\nactively sharing their location through a variety of applications and services.\nAlong with the continuing growth of Location-Based Social Networks (LBSNs),\nsecurity experts have increasingly warned the public of the dangers of exposing\nsensitive information such as personal location data. Most importantly, in\naddition to the geographical coordinates of the user's location, LBSNs allow\neasy access to an additional set of characteristics of that location, such as\nthe venue type or popularity.\n In this paper, we investigate the role of location semantics in the\nidentification of LBSN users. We simulate a scenario in which the attacker's\ngoal is to reveal the identity of a set of LBSN users by observing their\ncheck-in activity. We then propose to answer the following question: what are\nthe types of venues that a malicious user has to monitor to maximize the\nprobability of success? Conversely, when should a user decide whether to make\nhis/her check-in to a location public or not? We perform our study on more than\n1 million check-ins distributed over 17 urban regions of the United States. Our\nanalysis shows that different types of venues display different discriminative\npower in terms of user identity, with most of the venues in the "Residence"\ncategory providing the highest re-identification success across the urban\nregions. Interestingly, we also find that users with a high entropy of their\ncheck-ins distribution are not necessarily the hardest to identify, suggesting\nthat it is the collective behaviour of the users' population that determines\nthe complexity of the identification task, rather than the individual\nbehaviour.\n