Collaborative User Security Using Nearest Neighbor Approach in OSN

Online Social Network (OSN) connects users to share information from one to another or group. Information sharing process requires high security. But in OSN, information security and trusted users identification is in low level. Because information shared with friends of friends affects security policy mechanisms. This work uses a classification algorithm for identifying users who are sharing information securely in social networks based on some constraint and implement the policies to the group. The First Module of this work is to create new user and validate that user whether their profile is correct or not. Whenever user enters the login form an additional security mechanism is implemented. This system initiates Profile creation process that represents the user account creation process and content uploading process. After profile creation process friends list are categorized. Category may be based on user's neighbors, colleagues, etc. information are shared to one or more users specified in the category, based on the shared users list, this work identifies the specific category of users shared the information to more number of users. Nearest neighbor approach performs to identify the shared users category. Voting schemes are also used to identify shared users category. Threshold value is calculated from the voting schemes and compares the results for identifying the trusted users information shared in the network. In this, user may share their data to others in a secure manner and reduces the security issues in the social network.