Distributed Differential Privacy Algorithmand Application

The processing and analysis of big data brings convenience to people’s understanding of themselves and the world,but also brings the risk of sensitive data leakage.This paper focuses on the privacy protection of distributed storage data.Specifically,the research content of this paper includes the following two parts:(1)A distributed differential privacy Newtonian algorithm for big data processing and analysis is proposed.A distributed output disturbance Newton-type algorithm based on quasi-Newton method is proposed to realize distributed differential privacy by adding perturbation to the output result of the algorithm.Furthermore,in order to prevent possible local privacy leakage in the process of computer information interaction,a distributed gradient perturbation Newtonian algorithm is proposed by adding perturbation to the iterative process of the algorithm.The proposed algorithm can effectively reduce the communication cost of distributed system and improve the efficiency of privacy protection.Experiments show that the algorithm can analyze and process distributed storage data effectively and efficiently.(2)A distributed differential privacy data publishing algorithm is proposed in this paper.We analyze the privacy protection ability of the proposed algorithm.Secondly,we apply the algorithm proposed in Part(1)to published genetic data to conduct Genome-Wide Association Studies(GWAS).Experiments show that the algorithm proposed in this section can reduce and analyze distributed genetic data while protecting original data.Furthermore,the distributed machine learning algorithm for genetic data is studied.