【摘要】：差分隱私(Differential Privacy,DP)是一種新型的隱私保護(hù)技術(shù),它通過(guò)對(duì)原始查詢結(jié)果添加噪音后發(fā)布出去來(lái)達(dá)到隱私保護(hù)的目的。差分隱私下的隱私保護(hù)水平由隱私保護(hù)預(yù)算ε決定。針對(duì)以往的差分隱私數(shù)據(jù)發(fā)布后無(wú)法通過(guò)調(diào)整隱私保護(hù)預(yù)算以滿足用戶動(dòng)態(tài)對(duì)隱私保護(hù)水平變化的需求,論文提出了一種面向敏感數(shù)據(jù)逐步發(fā)布的差分隱私可調(diào)高斯機(jī)制。其主要研究?jī)?nèi)容包括:(1)提出可調(diào)高斯機(jī)制。證明了在統(tǒng)計(jì)數(shù)據(jù)的發(fā)布過(guò)程中,對(duì)原始數(shù)據(jù)添加服從高斯分布的噪音1V,發(fā)布滿足1 1(ε,δ)-差分隱私的結(jié)果1y,若將隱私保護(hù)預(yù)算1ε調(diào)整到2ε2 1(εε),在噪音1V的基礎(chǔ)上添加噪音2V,發(fā)布更精確的數(shù)據(jù)結(jié)果2y,則最終的發(fā)布結(jié)果1 2(y,y)滿足2 2(ε,δ)-差分隱私。(2)給出了可調(diào)高斯機(jī)制下數(shù)據(jù)發(fā)布的過(guò)程。采用可調(diào)高斯機(jī)制將隱私保護(hù)預(yù)算從1ε放大到2ε,并將隱私保護(hù)預(yù)算的一次放大擴(kuò)展到多次放大,在數(shù)據(jù)發(fā)布的過(guò)程中,若對(duì)原始統(tǒng)計(jì)數(shù)據(jù)添加具有馬爾可夫性質(zhì)的高斯噪音,即下一次添加多少噪音只與當(dāng)前添加噪音的多少有關(guān),與過(guò)去添加的噪音量無(wú)關(guān),則可以實(shí)現(xiàn)隱私保護(hù)預(yù)算的多次放大,從而達(dá)到差分隱私下敏感數(shù)據(jù)的逐步發(fā)布。(3)可調(diào)高斯機(jī)制的評(píng)估�；诓罘蛛[私下隱私損失的定義,比較了可調(diào)拉普拉斯機(jī)制、高斯組合機(jī)制、可調(diào)高斯機(jī)制等機(jī)制的隱私損失,發(fā)現(xiàn)采用可調(diào)高斯機(jī)制調(diào)整隱私保護(hù)預(yù)算的隱私損失小,驗(yàn)證了采用可調(diào)高斯機(jī)制放大隱私保護(hù)預(yù)算具有一定的優(yōu)越性。論文對(duì)可調(diào)高斯機(jī)制下的數(shù)據(jù)發(fā)布進(jìn)行了實(shí)驗(yàn)分析,采用該機(jī)制放大隱私保護(hù)預(yù)算的過(guò)程中,發(fā)布數(shù)據(jù)的均方誤差越來(lái)越小,即用戶獲得的數(shù)據(jù)結(jié)果精度越來(lái)越高。該研究面向一般意義的(ε,δ)-差分隱私,根據(jù)用戶的隱私需求調(diào)整隱私保護(hù)預(yù)算來(lái)發(fā)布數(shù)據(jù)的,使得隱私數(shù)據(jù)保護(hù)和使用之間更易于平衡,有利于推動(dòng)差分隱私的應(yīng)用,促進(jìn)大數(shù)據(jù)技術(shù)的發(fā)展。
[Abstract]:Differential privacy (Differential Privacy,DP) is a new privacy protection technology. It can achieve the purpose of privacy protection by adding noise to the original query results and publishing them out. The level of privacy protection under differential privacy is determined by the privacy protection budget 蔚. In view of the past differential privacy data can not be adjusted by adjusting the privacy protection budget to meet the needs of users to change the level of privacy protection, this paper proposes a differential privacy adjustable Gao Si mechanism for the gradual release of sensitive data. The main research contents are as follows: (1) the adjustable Gao Si mechanism is proposed. Proves that in the process of releasing statistical data, To add noise from Gao Si distribution to original data 1V, release the result that satisfies 11 (蔚, 未) -difference privacy 1y. if we adjust the privacy protection budget 1 蔚 to 2 蔚 21 (蔚), add noise 2V on the basis of noise 1V, release more accurate data result 2y. then The final result 12 (YY) satisfies 22 (蔚, 未) -difference privacy. (2) the process of data release under the adjustable Gao Si mechanism is given. The adjustable Gao Si mechanism is used to enlarge the privacy protection budget from 1 蔚 to 2 蔚, and the one-time enlargement of the privacy protection budget is extended to multiple magnification. That is, how much noise is added next time is only related to how much noise is added at present, and not related to the amount of noise added in the past, so that the privacy protection budget can be magnified several times. Thus, the sensitive data can be released step by step under differential privacy. (3) the evaluation of adjustable Gao Si mechanism. Based on the definition of privacy loss under differential privacy, the privacy loss of adjustable Laplace mechanism, Gao Si combination mechanism and adjustable Gao Si mechanism are compared. Verify that the adjustable Gao Si mechanism to enlarge the privacy protection budget has some advantages. This paper makes an experimental analysis on the data release under the adjustable Gao Si mechanism. In the process of enlarging the privacy protection budget by this mechanism, the mean square error of the published data becomes smaller and smaller, that is, the accuracy of the data obtained by the user is getting higher and higher. This research aims at (蔚, 未) -differential privacy in general sense, adjusts the privacy protection budget to release data according to the user's privacy needs, makes the balance between privacy data protection and use easier, and promotes the application of differential privacy. To promote the development of big data's technology.
【學(xué)位授予單位】：西北農(nóng)林科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP309

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王紅;葛麗娜;王麗穎;張靜;張翼鵬;;面向數(shù)據(jù)直方圖發(fā)布的差分隱私保護(hù)綜述[J];計(jì)算機(jī)應(yīng)用研究;2017年06期

2 穆海蓉;丁麗萍;宋宇寧;盧國(guó)慶;;DiffPRFs:一種面向隨機(jī)森林的差分隱私保護(hù)算法[J];通信學(xué)報(bào);2016年09期

3 張嘯劍;孟小峰;;基于差分隱私的流式直方圖發(fā)布方法[J];軟件學(xué)報(bào);2016年02期

4 何賢芒;王曉陽(yáng);陳華輝;董一鴻;;差分隱私保護(hù)參數(shù)ε的選取研究[J];通信學(xué)報(bào);2015年12期

5 張文靜;李暉;;差分隱私保護(hù)下的數(shù)據(jù)分級(jí)發(fā)布機(jī)制[J];網(wǎng)絡(luò)與信息安全學(xué)報(bào);2015年01期

6 朱天清;何木青;鄒德清;;基于差分隱私的大數(shù)據(jù)隱私保護(hù)[J];信息安全研究;2015年03期

7 王俊麗;管敏;魏紹臣;;面向社交網(wǎng)絡(luò)分析的差分隱私保護(hù)研究綜述[J];高技術(shù)通訊;2015年03期

8 丁麗萍;盧國(guó)慶;;面向頻繁模式挖掘的差分隱私保護(hù)研究綜述[J];通信學(xué)報(bào);2014年10期

9 孟小峰;張嘯劍;;大數(shù)據(jù)隱私管理[J];計(jì)算機(jī)研究與發(fā)展;2015年02期

10 張嘯劍;孟小峰;;面向數(shù)據(jù)發(fā)布和分析的差分隱私保護(hù)[J];計(jì)算機(jī)學(xué)報(bào);2014年04期

相關(guān)博士學(xué)位論文 前1條

1 李楊;差分隱私保護(hù)數(shù)據(jù)聚合優(yōu)化方法及其在數(shù)據(jù)可視化中的應(yīng)用[D];廣東工業(yè)大學(xué);2013年

相關(guān)碩士學(xué)位論文 前1條

1 王寶楠;基于差分隱私拉普拉斯機(jī)制的線性回歸分析研究[D];安徽理工大學(xué);2016年

，

本文編號(hào)：2259081