To connected with world online communication and sharing information using the social network sites become a very famous in recent days. But it’s very challenging job for social network site to provide the privacy and security. However, users need to become new friends to increase their social connections as well as to get information from specific group of people. Many online social networks (OSNs) using the past Friend recommendation method and which is very popular now days. There is a huge requirement to implement privacy-preserving friend recommendation methods for social networks as user privacy is the main motive nowadays. Online Social Networks(OSNs), not only get the focused from millions people to spend their time every day on social networks but also incredibly implement OSN clients social circles by using companion suggestions. This paper is motivated by need of friend proposal without showing privacy and security when using social networks. The goal in implementing of our system is to support users of OSN by securely creates trust with a stranger which is accomplishing by multi-hop recommendation process. Active OSN user privacy protection by using proposed method which is allowing them to enhanced their social networks. To carry out the secured social directed coordinating, existing framework use the secure kNN plan. Yet, with the help of KNN, distance based learning is not clear which sort of distance to use and their component to use to give the best results and calculation expense is very high. To overcome on this limitation and increased the outcomes precision, proposed framework utilizes SVM classifier for secure social coordinate matching. Through security analysis and trial outcomes, we demonstrate that the security, feasibility and precision of the proposed method is to have superior to anything existing one.

Online social network multi hop relationship trust value privacy.