A Study on Leakage of Private Information in Social Networks and its Preventive Measures

Social Networks are increasingly utilized by many people.These networks allow users to publish details about themselves.Some of the information provided by the users in the networks mean to be private information.Private information should be protected against unwarranted disclosure and only “Legitimate users” have a right to access it.It can be harmful to data owners and data users if it is misused.Although it is possible to use some types of learning algorithms on released data to predict private information. Inference attacks are initiated using released social networking data to predict private information. Collective inference methods are used to predict sensitive attributes of the data set.Network classification can be carried out with the combination of node details and connecting links in the graph model. Navie bayes classification is used to classify friendship links in a network in which local classifier, a relational classifier, and a collective inference methods are the three components used in the Network classification . Local classifier is a type of learning method and it can be applied in the initial step of collective inference. The relational classifier is a separate type of learning method and it analyzes the link structure and details of the node to identify a model for classification. Collective inference method can be used to increase the classification accuracy from the local and relational data details,which greatly reduces the accuracy of local classifier and give us the maximum accuracy through any combination