Privacy Adaptation for Secured Associations in a Social Cloud

Abstract

Mobile social networks' (MSN) diverse security concerns have immensely compromised users' personal details leaving them vulnerable to cybercrimes. This paper proposes an adaptive privacy architecture which provides content, identity and location privacy against disclosure of information that the user intends to keep private. The architecture implements multiple servers with the global social graph cached in the front-end server and sub-graph of each user across several servers. Content privacy is enforced through a role based access control model (RBAC) that relies primarily on the relationship and the trust between the users. Identity privacy is preserved through pseudonym generation. Location obfuscation is performed to safeguard the mobile user's location information. This ensures that the privilege and control is given to the user rather than a central authority namely social network service provider as suggested in the existing systems.