Research On Multi-Domain Data Privacy Protection Technology For Cloud Platform

With the rapid development and extensive use of cloud computing,more and more data are collected and stored in the cloud platform,which facilitates the use and sharing of data.Through data mining and other methods to analyze these data to bring convenience to users,but also for enterprises and organizations brought a lot of income.However,as the amount of data storage and processing is more and more big,through external data table identify the specific user data such as links to attack also often happen,this allows the user personal information is facing greater privacy risks,let us realize that it is necessary according to the relevant information security policy to protect user data privacy.In order to deal with the problem of data privacy disclosure in the process of data release and sharing,the technology of static and anonymous privacy protection data release for data record table is proposed,which limits the privacy disclosure risk in the process of data analysis to some extent,and at the same time guarantees the demand for normal data analysis.This paper studies the privacy protection of recorded phenotypic data in different fields on cloud platform to meet the specific needs of data analysis and transaction processing scenarios.For the research of privacy protection technology applied in the data analysis scenario,this paper mainly applies the privacy data release technology statically and dynamically from the aspects of stored procedure and query process.In the storage process,a data anonymization method that meets the k-anonymization principle is implemented to conduct static anonymization of data,and the anonymized data is stored in the cloud platform for data analysis to reduce the risk of data privacy information disclosure.In the process of query,implements a dynamic anonymous method,based on the query to limit the range of data available to the data analyst,and through the experimental analysis of the dynamic anonymous method in privacy and data availability,the effect of the experimental results show that,within a certain proportion of subset of data for dynamic desensitization,its good performance in two ways from traditional static elimination method.For applied in transaction processing scenario of privacy protection technology research,this paper focuses on the user has permissions using original data and personalized settings the need of privacy constraints,this paper proposes a method for recommended combination properties,to flexibly select part as quasi identifier attribute,in to protect the privacy of combination properties at the same time,better retain the other attributes of the information complete degree;In addition,the anonymous data is restored to the corresponding original data by establishing the corresponding relation between the original data set and the anonymous data set.Finally,this paper implements a prototype system based on the above privacy protection technologies and methods,which to some extent meets the needs of data analysts,users and cloud platform managers to analyze and use data while protecting data privacy information.