Research On Privacy Protection Method For Dynamic Network

In recent years,with the rapid development of Internet technology,we have entered the era of big data.Big data is a huge data set that relies on the Internet platform,and it has a large scale,rapid changes,low value density,and a wide variety.Features.In our daily lives,as various data spread on the Internet,there is a growing problem of leaking user privacy information.Companies or organizations of all sizes have experienced more or less malicious Internet attacks such as applications and Trojans cause their user data to flow out.Therefore,the issue of privacy protection has been called a hot issue that many research scholars pay attention to.Most of the existing research on privacy protection is aimed at static data in static social networks.However,in fact,the data we publish in social networks is changing all the time,and is closer to dynamic social networks.For the large-scale and fast-flowing streaming data in dynamic social networks,the current method is only the privacy protection algorithm that directly applies static data,so that although the user privacy can be well protected at a certain time,once the attacker utilizes Publishing data in multiple time slices for joint attacks makes it easy to expose private information.In view of the above problems,this paper proposes a privacy protection model for dynamic networks.Firstly,an active risk detection mechanism is proposed.Based on the analysis of the advantages of Bayesian network in analyzing the uncertainty problem and predicting the state of the dynamic social network at the next moment,the probabilistic model is used to estimate the probability of user privacy leakage and decide whether to take it.Privacy protection measures;Secondly,a collaborative privacy protection algorithm is adopted for data that cannot be determined for security.According to the length and sensitivity of the data distance,the sensitive information of the user is divided into data privacy space,and the data in the data privacy space is adopted.The related typical analysis algorithm finds the substitute value of the sensitive data.Finally,in order to improve the efficiency of the collaborative privacy protection algorithm,each time the data is calculated,the similarity of the data items under the adjacent time slice is measured,so that the weight of each label of the data item can be determined.Adaptive adjustment of values to achieve fast calculation of privacy protection parameter values.Based on the proposed privacy protection model,a detailed,systematic design scheme and specific algorithm are proposed,and the feasibility of the experiment is designed.Experiments show that the social network after the algorithm is processed has less loss of structural information.