Research On Privacy Preservation Technology Based On Clustering

With the deployment and implementation of the national big data strategy,the seemingly massive mass network data is endowed with new application value and research significance through data mining technology.However,the data generation and collection capacity of a few companies or research institutes can not meet the data requirement of data mining research.Therefore,Internet data sharing and publishing has become an important basic step in the research of big data.If it is not appropriate to protect users' privacy data in the network,but directly share in the network,it is very likely to be exploited by illegal elements.It leads to the leakage of the user's privacy,the loss of the user's economy,even the threat of public security and the damage to the interests of the country.According to the above situation,the privacy preservation technology facing the release of data has important research value and research significance.This paper studies the existing privacy protection technology,compare the advantages and disadvantages.The anonymous technology is widely used in various fields due to its simple operation,high feasibility and good privacy preservation.The classic k-anonymous model is widely used,and the researchers carries on the improvements and innovations.In this paper,we study privacy preservation technology based on clustering.At first,this paper combines the kmeans clustering and cuckoo search algorithm,and the design objective function to quickly select the initial cluster center;Secondly,this paper proposes(k,t)anonymous model:(1)the equivalence class has at least k records.(2)the distance between the sensitive attribute distribution in the equivalence class and the whole dataset is not larger than the threshold t,in order to effectively enhance the strength of privacy preservation.Finally,this paper uses the C++ language to implement the proposed algorithm and model,and uses the standard data set Adults to verify the proposed algorithm performance improvement than the existing methods.privacy protection technology facing to release data has important research value and research significance.