Research On Privacy Protection Method For Sensitive Relationship Social Network

In recent years, with the rapid development of internet technology and information technology,a large number of social network continually raise,more and more users information is exposed in the social network. If these information are not properly used,and it will cause problems to people’s life and may even pose a serious threat. As a result, people are paying more attention to the privacy informationprotected effectively in the social network, the network users’ privacy information security is very important.In the study of social network privacy protection technology,the safe and effective anonymous methods become the most commonly used method, and have become the hotspot in the current study. The main idea of theanonymous methodis that the original social network makes some transformation to protect the individual privacy information on the social network. Most of the existing social network privacy protection anonymous methods only for a single type ofindividual information on social network, kinds of social network in real life are also very common, and kinds of social networks in the sensitive relationship cannot be effectively protected.On the basis of checking and organizinga large number of literature at domestic and overseas, this thesis studies the privacy protection problem on social network for the multi-type relationship.This thesisimproved k- anonymous model and build the(k2,l)- anonymous model the(k2,l)- anonymous model, the anonymous model is put forward three requirements, can also stop sensitive relationship attack,friend connect attack and node degrees attack. The main research contents of thesis include the following three aspects:(1) This thesis introduced the research status at home and abroad on social network privacy protection method, the existing anonymous methodshas carried onthe preliminary understanding. This thesis introduces the way of privacy and common attack forms in details. Then,systematically introduces the anonymous method on social network, summarized in detail and summarizes the basic k- anonymous model and several kinds of commonly used anonymous models based on k- anonymous improved in details, and different anonymous models for different types of background knowledge of the attack have different advantages and disadvantages.(2)In order to ensure that users’ privacy on social network for sensitive relationship will not be attacked by the existence of multiple types of attackersat the same time. This thesisimproved k- anonymous model and build the(k2,l)-anonymous model the(k2,l)- anonymous model, because of the intensity of privacy protection and social network diagram need to be balanced between availability, while the privacy protection strength enhancement means that the availability reduced. In order to meet the demand of privacy protection at the same time as much as possible to improve the availability of social network, the first, according to the rule of adding sensitiveedge to social networks, the thesis add edge to the social network, this thesis use the dynamic planning sequence algorithm and greedy algorithm anonymous degrees sequence algorithm for grouping nodes, and build the(k2, l)- anonymous model.(3)In order to verify the feasibility and effectiveness of the anonymous model, this thesis make the simulation experiments on the datasets, through the comparison and analysis oftwo implementation(k2, l)- anonymous model algorithms by setting different parameters.In this thesis, through the experimental simulation, this improved the feasibility and effectiveness of the anonymous algorithm and anonymous model, users can according to their different needs, depending on the values of k and l value choose different anonymous algorithm, in ensuring the good condition of social network diagram and theusers’ satisfactory.