TY - GEN
T1 - It's the way you check-in: Identifying users in location-based social networks
AU - Rossi, Luca
AU - Musolesi, Mirco
PY - 2014/10/1
Y1 - 2014/10/1
N2 - In recent years, the rapid spread of smartphones has led to the increasing popularity of Location-Based Social Networks (LBSNs). Although a number of research studies and articles in the press have shown the dangers of exposing personal location data, the inherent nature of LBSNs encourages users to publish information about their current location (i.e., their check-ins). The same is true for the majority of the most popular social networking websites, which offer the possibility of associating the current location of users to their posts and photos. Moreover, some LBSNs, such as Foursquare, let users tag their friends in their check-ins, thus potentially releasing location information of individuals that have no control over the published data. This raises additional privacy concerns for the management of location information in LBSNs. In this paper we propose and evaluate a series of techniques for the identification of users from their check-in data. More specifically, we first present two strategies according to which users are characterized by the spatio-temporal trajectory emerging from their check-ins over time and the frequency of visit to specific locations, respectively. In addition to these approaches, we also propose a hybrid strategy that is able to exploit both types of information. It is worth noting that these techniques can be applied to a more general class of problems where locations and social links of individuals are available in a given dataset. We evaluate our techniques by means of three real-world LBSNs datasets, demonstrating that a very limited amount of data points is sufficient to identify a user with a high degree of accuracy. For instance, we show that in some datasets we are able to classify more than 80% of the users correctly.
AB - In recent years, the rapid spread of smartphones has led to the increasing popularity of Location-Based Social Networks (LBSNs). Although a number of research studies and articles in the press have shown the dangers of exposing personal location data, the inherent nature of LBSNs encourages users to publish information about their current location (i.e., their check-ins). The same is true for the majority of the most popular social networking websites, which offer the possibility of associating the current location of users to their posts and photos. Moreover, some LBSNs, such as Foursquare, let users tag their friends in their check-ins, thus potentially releasing location information of individuals that have no control over the published data. This raises additional privacy concerns for the management of location information in LBSNs. In this paper we propose and evaluate a series of techniques for the identification of users from their check-in data. More specifically, we first present two strategies according to which users are characterized by the spatio-temporal trajectory emerging from their check-ins over time and the frequency of visit to specific locations, respectively. In addition to these approaches, we also propose a hybrid strategy that is able to exploit both types of information. It is worth noting that these techniques can be applied to a more general class of problems where locations and social links of individuals are available in a given dataset. We evaluate our techniques by means of three real-world LBSNs datasets, demonstrating that a very limited amount of data points is sufficient to identify a user with a high degree of accuracy. For instance, we show that in some datasets we are able to classify more than 80% of the users correctly.
KW - Location-based social networks
KW - Privacy
KW - User identification
UR - http://www.scopus.com/inward/record.url?scp=84912127066&partnerID=8YFLogxK
U2 - 10.1145/2660460.2660485
DO - 10.1145/2660460.2660485
M3 - Conference article published in proceeding or book
AN - SCOPUS:84912127066
T3 - COSN 2014 - Proceedings of the 2014 ACM Conference on Online Social Networks
SP - 215
EP - 225
BT - COSN 2014 - Proceedings of the 2014 ACM Conference on Online Social Networks
PB - Association for Computing Machinery, Inc
T2 - 2nd ACM Conference on Online Social Networks, COSN 2014
Y2 - 1 October 2014 through 2 October 2014
ER -