A Closer Look: Evaluating Location Privacy Empirically

The breach of users’ location privacy can be catastrophic. To provide users with privacy protection, numerous location privacy methods have been developed. While several literature surveys exist in this field, the lack of comparative empirical evaluations imposes challenges for adopting location privacy by applications and researchers in a wide range of domains. This talk presents our recent study which fills the gap by evaluating location privacy with real-world datasets. For utility evaluation, we consider various types of measures, such as distortion-based and count-based measures, as well as individual's mobility patterns; for privacy protection evaluation, we design two empirical privacy risk measures via inference and re-identification attacks. Furthermore, we study the computational overheads incurred by location privacy. The results show that it is possible to strike a balance between utility and privacy when sharing location data with untrusted servers.