Privacy Preservation in Sequential Published Social Networks Against Mutual Friend Attack

In recent years the social networks are widely used the way of connecting people, interact with each other and share the information. The social network data is rich in content and the data are published for third party users such as researchers. The social interaction between individual’s changes rapidly as time changes so there is a need of privacy preserving in dynamic networks. An adversary can acquire some local knowledge about individuals in the network and can easily breach the privacy of a few victims. This paper mainly focuses on preserving privacy in sequential published network data where the adversary has some knowledge about the number of mutual friends of the target victims over a time period. The kw-Number of Mutual Friend Anonymization model is proposed to anonymize each sequential published network. In this privacy model, k indicates the privacy level and w is the time interval taken by the adversary to acquire the knowledge of the victim. By this approach the adversary cannot identify the victim by acquiring the knowledge of each sequential published data. The performance evaluation shows that the proposed approach can preserve many characteristics of the dynamic social networks.