Privacy-preserving Predictive Service Mechanism In Big Data Environment

With the rapid development of Internet technology,big data has been integrated into all aspects of people's life,big data can provide people with more accurate and more efficient decision-making support.As one of the important applications of big data,predict service has been widely used in the fields of disease diagnosis,financial analysis and malware detection in recent years and has attracted great interest from researchers.However,the widespread use of big data not only brings great convenience to people's life,but also a huge threat to people's privacy at the same time.In recent years,due to a variety of privacy disclosure events,people are paying more and more attention to the personal privacy issue.The privacy protection issue in the predictive service in big data environment mainly focuses on the following three aspects: the privacy protection of the user query information,the protection of the service provider's service model and the security protection of the data in the service collection.In order to solve the above problems,a privacy preserving predictive service scheme based on the nave Bayesian classification algorithm is proposed in this dissertation and then we apply it to a medical diagnosis scene in big data environment,and implement a secure and efficient medical pre-diagnosis service system.Specifically,this dissertation has done the following aspects of the work:(1)In order to meet the specific requirements of predictive service in big data environment,we improve the classical Naive Bayesian classification algorithm and apply it to the privacy-preserving classification predictive scheme in big data environment.(2)In order to protect users' personal information and service providers' service model,this dissertation constructs a privacy-preserving predictive scheme by using random hiding and polynomial aggregation technology.Security and efficient data transmission between the data collector and the data provider are guaranteed by introducing a secure symmetric encryption scheme and data checking mechanism.The correctness,security and calculation complexity of the scheme are analyzed in detail.The scheme proposed has better security and higher efficiency.(3)A security medical pre-diagnosis system is implemented by applying the proposed scheme to the medical pre-diagnosis service scene.The system consists of a mobile client and a predicting server.It can ensure that the online medical pre-diagnosis service provider can collect medical diagnosis instances securely and ensure the privacy and security of data in the process of collecting.On the other hand,the system provide users who need medical pre-diagnosis with a complete set of security services,including registration,registration,medical pre-diagnosis,diagnostic history management,key updates and so on.(4)The medical pre-diagnostic system designed in this paper was implemented.In order to evaluate the function and performance of each module of the system designed in this dissertation,the actual implementation of the system is realized.A large number of simulation results show that the system realized all the functions designed and can serve many users at the same time.