What Are You Willing to Sacrifice to Protect Your Privacy When Using a Location-Based Service?

Abstract

Today, we use location-based services on a daily basis. They provide information related to the current location of the users and are extremely helpful. The next step of location-based services is to use predicted locations of users to create new content or to improve the quality of existing ones. Today location-based services must capture a large location history of the users in order to be able to build predictive mobility models of users and forecast their future locations. This is a clear privacy issue because these services are thus able to also obtain sensitive information related to users. In this paper, we propose a system that provides future locations of users to location-based services, protects the location privacy of the users with spatio-temporal conditions and ensures the utility of the information provided by the location-based services. The user is a key actor of the system and is involved in the protection process because she indicates these spatio-temporal conditions, which express the spatio-temporal utility she is willing to sacrifice in order to protect her privacy. We evaluate the two components of the system according to two perspectives: a prediction accuracy analysis and a utility/location privacy evaluation. The proposed system provides satisfactory prediction accuracy results that exceed 70%. The utility/privacy evaluation shows that our mechanism obtains the best results in terms of utility and location privacy compared to two other common location privacy preserving mechanisms. Hence, these evaluations confirm the relevance of our system.