Privacy Preservation of Social Network Data against Structural Attack using K-Auto restructure

Abstract — Recent few years social networking sites are facing a problem of various attacks on their network which contains sensitive data. Usually social networks will publish their social network data for research purpose. Researchers and social network analysts can make use of these data to do research for decision making and market analysis. Before releasing the data for research, the social network site removes the identifiable parameters such as name, location, type of relationship, etc. Simply removing all identifiable personal information before releasing the data is insufficient. It is easy for an adversary to identify the target by performing different structural queries. Many of the previous studies were concentrated only on the anonymization part. We identify a special type of attack called structural attack. With the aim of resisting various structural attacks, in this paper, we proposed a new and efficient framework called k-Autorestructure which to protect against multiple structural attacks. There is no doubt in that our proposed algorithm will resist any kind of structural attack again the social network data