Research on Improvement of Bayesian Network Privacy Protection Algorithm Based on Differential Privacy

doi:10.3969/j.issn.1671-1122.2020.11.010

Abstract

Abstract:

In response to the urgent need for desensitization protection algorithms by the data companies and open government publishing data, under the strict differential privacy theory, an improved Bayesian network algorithm FCPrivBayes with an attribute segment preference mechanism and a clustering algorithm is proposed, which avoids the random selection of the attributes of the first attribute segment, and uses the clustering method to replace the equal-width method to discretize the data. Experimental data show that FCPrivBayes effectively improves data utility indicators while ensuring the data privacy protection effect. Which provides new technical options for data companies to protect data and for government to release data, and benefits the user privacy protection and the development of the big data industry.

Key words: differential privacy theory, Bayesian network algorithm, privacy protection

CLC Number:

TP309

XIAO Biao, YAN Hongqiang, LUO Haining, LI Jucheng. Research on Improvement of Bayesian Network Privacy Protection Algorithm Based on Differential Privacy[J]. Netinfo Security, 2020, 20(11): 75-86.

Figures/Tables 11

References 17

[1]	ZHANG Xiaojian, MENG Xiaofeng. Differential Privacy in Data Publication and Analysis[J]. Chinese Journal of Computers, 2014,37(4):927-949.
	张啸剑, 孟小峰. 面向数据发布和分析的差分隐私保护[J]. 计算机学报, 2014,37(4):927-949.
[2]	LI Haoran, XIONG Li, ZHANG Lifan, el al. DPSynthesizer: Differentially Private Data Synthesizer for Privacy Preserving Data Sharing[J]. Proceedings of the VLDB Endowment, 2014,7(13):1677-1680. doi: 10.14778/2733004.2733059 URL pmid: 26167358
[3]	OUYANG Jia, YIN Jian, LIU Shaopeng. Differential Privacy Publishing Strategy for Distributed Transaction Data[J]. Journal of Software, 2015,26(6):1457-1472.
	欧阳佳, 印鉴, 刘少鹏. 一种分布式事务数据的差分隐私发布策略[J]. 软件学报, 2015,26(6):1457-1472.
[4]	ZHANG Lin, LIU Yan, WANG Ruchuan. Location Publishing Technology Based on Differential Privacy-preserving for Big Data Services[J]. Journal on Communications, 2016,37(9):46-54.
	张琳, 刘彦, 王汝传. 位置大数据服务中基于差分隐私的数据发布技术[J]. 通信学报, 2016,37(9):46-54.
[5]	HARICHANDAN R, MURAT K, LATANYA S. Practical Differentially Private Modeling of Human Movement Data[EB/OL]. http://link.springer.com/chapter/10.1007/978-3-319-41483-6_13, 2020-7-16.
[6]	XIAO Xiaokui, WANG Guozhang, GEHRKE Johannes. Differential Privacy via Wavelet Transforms[J]. IEEE Transactions on Knowledge & Data Engineering, 2011,23(8):1200-1214.
[7]	XIAO Xiaokui, WANG Guozhang, GEHRKE , et al. Differential Privacy via Wavelet Transforms[J]. IEEE Transactions on Knowledge & Data Engineering, 2011,23(8):1200-1214.
[8]	ZHANG Jun, GRAHAM C, CECILIA M, et al. PrivBayes: Private Data Release via Bayesian Networks[EB/OL]. https://www.researchgate.net/publication/320679178_PrivBayes_Private_Data_Release_via_Bayesian_Networks, 2020-7-16.
[9]	LI Mingzhu. Privacy Data Publishing Method Based on Smoothness Sensitivity in Bayesian Networks[D]. Inner Mongolia University, 2019.
	李明珠. 基于平滑敏感度的贝叶斯网络隐私数据发布方法[D]. 内蒙古大学, 2019.
[10]	HAO Zhifeng, WANG Riyu, CAI Ruichu, et al. Privacy Data Publishing Method Based on Bayesian Network and Semantic Tree[J]. Computer Engineering, 2019,45(4):124-129.
	郝志峰, 王日宇, 蔡瑞初, 温雯. 基于贝叶斯网络与语义树的隐私数据发布方法[J]. 计算机工程, 2019,45(4):124-129.
[11]	ZHANG Xiaojian, CHEN Li, JIN Kaizhong, et al. Private High-dimensional Data Publication with Junction Tree[J]. Journal of Computer Research and Development, 2018,55(12):2794-2809.
	张啸剑, 陈莉, 金凯忠, 等. 基于联合树的隐私高维数据发布方法[J]. 计算机研究与发展, 2018,55(12):2794-2809.
[12]	WANG Liang, WANG Weiping, MENG Dan. Privacy Preserving Data Publishing via Weighted Bayesian Networks[J]. Journal of Computer Research and Development, 2016,53(10):2343-2353.
	王良, 王伟平, 孟丹. 基于加权贝叶斯网络的隐私数据发布方法[J]. 计算机研究与发展, 2016,53(10):2343-2353. doi: 10.7544/issn1000-1239.2016.20160465 URL
[13]	MCSHERRY F. Privacy Integrated Queries: An Extensible Platform for Privacy-preserving Data Ana1ysis[EB/OL]. https://dl.acm.org/doi/pdf/10.1145/1559845.1559850, 2020-7-16.
[14]	DWORK C, MCSHERRY F, NISSIM K, et al. Calibrating Noise to Sensitivity in Private Data Analysis[C]// Springer. Proceedings of the 3th Theory of Cryptography Conference (TCC), March 3-4, 2006, New York,USA. Heidelberg: Springer, 2006: 363-385.
[15]	MCSHERRY F, TALWAR K. Mechanism Design via Differential Privacy[EB/OL]. https://www.researchgate.net/publication/4294387_Mechanism_Design_via_Differential_Privacy, 2020-7-16.
[16]	BARAK , BOAZ C, DWORK K, et al. Privacy, Accuracy,Consistency too: A Holistic Solution to Contingency Table Release[EB/OL]. https://dblp.uni-trier.de/pid/83/6616.html, 2020-5-18.
[17]	ZHOU Zhihua. Machine Learning[M]. Beijing: Tsinghua University Press, 2016.
	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.

姓名	是否有糖尿病（X）
Ross	1
Monica	1
Joey	0
Phoebe	0
Chandler	1

疾病名称（Disease）	可用性
癌症（Cancer）	50
艾滋病（HIV）	20
乳头瘤病毒（HPV）	30

疾病名称（Disease）	exp($\varepsilon $q/2S(q))	概率
癌症（Cancer）	exp(25)	0.9999
艾滋病（HIV）	exp(10)	3?10^-7
乳头瘤病毒（HPV）	exp(15)	0.00004

疾病名称（Disease）	$\varepsilon $=0	$\varepsilon $=0.1	$\varepsilon $=0.5
癌症（Cancer）	0.333	0.628	0.993
艾滋病（HIV）	0.333	0.124	0.006
乳头瘤病毒（HPV）	0.333	0.231	0.001

$i$	${{A}_{i}}$	${{\text{ }\!\!\Pi\!\!\text{ }}_{i}}$
1	age	$\phi $
2	education	$\left\{ age \right\}$
3	workclass	$\left\{ age,education \right\}$
4	title	$\left\{ age,workclass \right\}$
5	income	$\left\{ workclass,title \right\}$