Big Data's Statistical Relevance Scheme For Privacy Protection And Its Application In Electronic Transactions

As more and more activities are carried out through computer networks,the amount of sensitive data stored by governments and other organizations is increasing.Many organizations hope to get more favorable information from the cooperation of their data,but the existence of privacy prevent the organizations from mining more data,and multi-party secure statistical analysis of data emerges as the times require.At the same time,electronic transactions overcome the limitations of time and geographical location,and greatly facilitate people's lives.Statistical analysis of consumer data in electronic transactions can also facilitate the provision of better services to consumers.Aiming at the security problems such as easily exposing personal privacy in big data's environment,this paper studies the multi-party secure big data statistical correlation analysis and completes its application in electronic transactions.1.We propose a correlation analysis scheme for privacy protection in big data environment,which collects and blocks a large number of users' data and transmits it to multiple servers through secure channels.Data synchronization storage in ciphertext processing process can effectively avoid internal attacks,so long as one of the servers is not breached,the system is secure.The Paillier homomorphic encryption algorithm is used to encrypt the data and the RSA algorithm is used to control the access authority.It realizes many kinds of correlation analysis and processing methods under ciphertext,including partial correlation,semi-partial correlation and complex correlation,and protects users' personal privacy information.2.We add the ElGamal homomorphic encryption algorithm,and construct a multi-party security complex correlation analysis for protecting the privacy of the consumers in the electronic transaction.The scheme solves the problem of security in the actual electronic transaction.We have improved the elgamal algorithm,implemented multi-party computing,and the servers and analysts who participate in the protocol have their own private keys,without which either party will not be able to obtain the analysis data.The scheme realizes multi-party security analysis of complex correlation.No participant can know the personal information of other participants and no trusted third party is needed.The scheme has very strong practicability in the electronic transaction,greatly reduces the risk of consumer data leakage,and can safely provide the data to the data analysis and improve the efficiency in any party of the electronic transaction.3.This paper summarizes the proposed scheme and deeply studies the multi-party security attack-related scheme in big data environment and its application in electronic transactions.The correctness and security of schemes are analyzed,and the performance of each scheme is compared in detail.