摘要
大数据中心的数据量巨大,存储方法通常为分布式存储,导致数据保护方法相对简单,黑客易利用相关漏洞窃取关键信息。且大部分数据隐私保护技术是针对单向静态数据的,因此如何有效地应对大数据动态风险问题是当前最为关键的问题之一。提出基于RSA公钥加密的大数据安全性双向检测方法。设计RSA公钥加密机制,通过用户的安全需求度选择不同的数据安全度,以满足各个等级的大数据安全需求。基于分类算法评价指标,引入动态选择算法,并将评价指标聚类成簇,利用精确度选择各个簇中的最优算法。确定权重取值,通过距离函数将不同权值组合处理形成组合权值,将其代入加权投票算法,最终获取大数据安全性双向检测结果。实验结果证明:对多类别大数据的安全性双向检测,研究方法的检测率高达98%,且误报率低于0.4%,约登指数接近0.9,以上所得实验数据均表明研究方法具有应用优势,实用性更强。
Due to massive data in the big data center,the existing method of data protection is too simple,and hackers are easy to use relevant vulnerabilities to steal key information.In addition,most data privacy protection technologies are only aimed at one-way static data.In order to effectively deal with the dynamic risk of big data,this paper presented a two-way detection method for big data security based on RSA public key encryption.Firstly,we designed an RSA public key encryption mechanism,and then chose different degrees of data security according to the users’demands,thus meeting the security needs of big data at all levels.Based on the evaluation index of the classification algorithm,we introduced a dynamic selection algorithm and clustered the evaluation indexes.Secondly,we used the accuracy to select the best algorithm and thus determine the weight value.Moreover,we used distance functions to combine different weights to form a combined weight,and then put it into the weighted voting algorithm.Finally,we obtained the two-way detection result.Experimental results show that the detection rate of the method is as high as 98%for the two-way detection of multi-class big data,and the false positive rate is less than 0.4%.The Yoden index is close to 0.9.The experimental data prove the strong practicability of the method.
作者
高丽杰
薛丽香
马照瑞
李振峰
GAO Li-jie;XUE Li-xiang;MA Zhao-rui;LI Zhen-feng(College of Information Engineering,Zhengzhou University of Science and Technology,Zhengzhou Henan 450064,China;Zhengzhou University of Light Industry,Zhengzhou Henan 450001,China;Zhengzhou University of Science and Technology,Zhengzhou Henan 450064,China)
出处
《计算机仿真》
北大核心
2023年第2期410-414,共5页
Computer Simulation
基金
河南省科技厅科技攻关项目(182102210552)
河南省高等学校重点科研项目(教科技〔2020〕348)
信息化助力高校教育治理能力与治理体系现代化研究(21A630035)。
关键词
公钥加密
大数据安全性
双向检测
组合加权
Public key encryption
Big data security
Two-way detection
Combined weighting
