Community-Structured Evolutionary Game for Privacy Protection in Social Networks
Open Access
- 2 October 2017
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 13 (3) , 574-589
Abstract
Social networks have attracted billions of users and supported a wide range of interests and practices. Users of social networks can be connected with each other by different communities according to professions, living locations, and personal interests. With the development of diverse social network applications, academic researchers, and practicing engineers pay increasing attention to the related technology. As each user on the social network platforms typically stores and shares a large amount of personal data, the privacy of such user-related information raises serious concerns. Most research on privacy protection relies on specific information security techniques such as anonymization or access control. However, the protection of privacy depends heavily on the incentive mechanisms of social networks, like users' psychological decisions on security execution and socio-economic considerations. For example, the desire to influence the behaviors of other people may change a user's choice of security setting. In this paper, a game theoretic framework is established to model users' interactions that influence users' decisions as to whether to undertake privacy protection or not. To model the relationship of user communities, community-structured evolutionary dynamics are introduced, in which interactions of users can only happen among those users who have at least one community in common. Then the dynamics of the users' strategies to take a specific privacy protection or not is analyzed based on the proposed community structured evolutionary game theoretic framework. Experiments show that the proposed framework is effective in modeling the users' relationships and privacy protection behaviors. Moreover, results can also help social network managers to design appropriate security service and payment mechanisms to encourage their users to take the privacy protection, which can promote the spreading of privacy behavior throughout the network.Keywords
Funding Information
- National Natural Science Foundation China (61371079, 61571300)
- U.S. Army Research Office (W911NF-16-1-0448)
- U.S. National Science Foundation (CMMI-1435778)
This publication has 38 references indexed in Scilit:
- A Novel Time-Obfuscated Algorithm for Trajectory Privacy ProtectionIEEE Transactions on Services Computing, 2013
- Understanding Social Networks From a Multiagent PerspectiveIEEE Transactions on Parallel and Distributed Systems, 2013
- Spreading of cooperative behaviour across interdependent groupsScientific Reports, 2013
- A survey of privacy in multi-agent systemsThe Knowledge Engineering Review, 2013
- A 61-million-person experiment in social influence and political mobilizationNature, 2012
- A review of evolutionary graph theory with applications to game theoryBioSystems, 2012
- Social networks and cooperation in hunter-gatherersNature, 2012
- Evolutionary dynamics in set structured populationsProceedings of the National Academy of Sciences, 2009
- The different limits of weak selection and the evolutionary dynamics of finite populationsJournal of Theoretical Biology, 2007
- Emergence of cooperation and evolutionary stability in finite populationsNature, 2004