Research Of The Privacy Protection Based On Dynamic Data Publication

With the rapid development of the information age, the increase in the amount ofinformation online, privacy protection have aroused widespread concern, data miningand data analysis has been widely applied. Government, enterprises and individuals asneeded on the Internet to publish their own data, these data contain a variety ofinformation in society, some involving sensitive information. After human fleshsearch network events, make everyone’s privacy protection awareness graduallystrong, they do not want their privacy information is leaked. Privacy research in therelease of the data on the individual, the industry and even society as a whole, have animportant significance. Sometimes the need to extract some information to analyzecurrent trends or specific characteristics of the population, ensuring personalinformation is not leaked premise and the effective implementation of data analysis isone of the elements of privacy protection. privacy protection policy are based onone-published data in the current, due to the rapid pace of data updates, they need toupdate database information simultaneously, so more consideration are needed in theprocess o f data re-released. If this problem can’t be solved, it will affect thedevelopment of other related technologies. To protect the privacy of sensitiveinformation protection has an important theoretical and practical value, privacyprotection for dynamic data released is a research focus of the researchers in recentyears.First, understanding overview of privacy protection, analysis is based on apublished privacy protection policy, understanding the probable outside attacks in thedata publication process about private information. For that, privacy protection policyis how to avoid privacy disclosure. After the analysis, privacy points are grasped,familiar the privacy protection ideas in theory.The work is done based on the study of privacy protection based on dynamicdata released. The main task is the privacy protection policies and anonymous methodtwo aspects. To study a privacy policy to guarantee the security of sensitive data in there-release of the data, such as a data set in add, delete and modify data. Taking intoaccount the current development trend of anonymity is to find ways to reduce the lossof information in the anonymous process. If the original privacy protection methodsare applied directly to the dynamic re-released of the data, that will lead to a lot ofprivate information leakage, needing to combine the re-release characteristics. theclassic m-invariance privacy policies, as a example, is analyzed and studied. Thelimitations and shortcomings of the m-invariance is the breaking point, put forward am-inclusion privacy protection policy supporting data modification and an anonymousmethod based on clustering.In the anonymous process, it adopts clustering thought, reducing the loss of information and ensuring the privacy of information security and usability. Comparem-invariance with m-inclusion through experiments, which prove that m-inclusiongood at degree in information loss and counterfeit number. And in order to measurethe m-inclusion strategy operating efficiency, record real data sets issued severalanonymous execution time, certify that strategy’s practicality.and also doing someanalysis in privacy security, in summary, m-inclusion is more satisfied the demand fordynamic data released.