With the sharing of pictures via web-based networking media, for example, Facebook, twitter, and so forth increments, keep up their security turns into the significant issue. As client shares their private pictures on social destinations, individuals anticipate that more apparatuses will permit them to recover control over their protection. By considering this need, we propose an Adaptive Privacy Policy Prediction (A3P) framework which gives client advantageous security settings via consequently producing customized arrangements. To characterize clients' protection inclinations we consider the distinctive elements for example, social environment, individual qualities, picture substance and metadata. For the pictures being transferred, we characterize the best accessible security arrangement for the client in light of the clients' accessible history on the site. For that we propose a two level system. A3P framework depends on the picture characterization structure for picture classes which might be connected with comparative arrangements and on a strategy forecast calculation to naturally create a strategy for each recently transferred picture, likewise as indicated by clients' social components