In Current Time, Social Network has become one of the most Powerful Resources for people to connect to their friends and family members or etc. and also publish their sensitive Information. Some of the information in social network is meant to be private. For different work, the Social Network Data like Medical Dataset or etc was given by the social sites to the advisor for some advertisement of products or etc. But Sometime, these datasets was used to infer the private information using various inference attacks. Here, we present a brief review of the inference attacks on social network dataset. We also identify the new challenges in privacy preserving on social network data comparing to the extensively studied relational case, and examine the possible problem. We survey the sanitization methods which used to preventing inference attacks.