Incentive Mechanism for Social Network Data Pricing under Privacy Preservation

Online social networks have become very important sources of personal data that contain preferences, tastes, interests and friendships of their users. The potential that these data may be exploited poses a rising concern over how privacy is protected by these social network platforms. As a consequence, users are starting to demand privacy protection and privacy compensation when their data are used. This situation begs the question: How to properly compensate social network users for the disclosure of their data hosted in the network while preserving their privacy? In this paper, we consider data trade between a large amount of privacy-aware social network users, and a data broker with a budget who wants to find out aggregate friendship information of this social network. We propose an incentive mechanism for pricing social network data and privacy preservation. We prove that the proposed mechanism satisfies many desirable properties, including incentive compatibility, individual rationality, budget balance, and node differential privacy. Further, the accuracy of the proposed mechanism is theoretically analysed and its effectiveness is validated by experiments on real-world datasets.