MAXGDDP：基于差分隐私的决策数据发布算法

doi:10.11959/j.issn.1000-436x.2018049

通信学报 ›› 2018, Vol. 39 ›› Issue (3): 136-146.doi: 10.11959/j.issn.1000-436x.2018049

MAXGDDP：基于差分隐私的决策数据发布算法

傅继彬¹,张啸剑¹,丁丽萍²

¹ 河南财经政法大学计算机与信息工程学院，河南郑州 450046
² 中国科学院软件研究所，北京 100190

修回日期:2018-02-27 出版日期:2018-03-01 发布日期:2018-04-02
作者简介:傅继彬（1975-），男，河南许昌人，博士，河南财经政法大学副教授，主要研究方向为知识工程、机器学习、隐私保护等。|张啸剑（1980-），男，河南周口人，博士，河南财经政法大学副教授，主要研究方向为隐私保护、差分隐私、数据库等。|丁丽萍（1965-），女，山东青州人，中国科学院软件研究所研究员、博士生导师，主要研究方向为数字取证、系统安全、可信计算等。
基金资助:
国家自然科学基金资助项目(61502146);国家自然科学基金资助项目(91646203);国家自然科学基金资助项目(91746115);河南省自然科学基金资助项目(162300410006);河南省科技攻关基金资助项目(142102210384);河南省科技攻关基金资助项目(172102310713);河南省教育厅高等学校重点科研基金资助项目(16A520002);河南省青年骨干教师基金资助项目;河南财经政法大学青年拔尖人才资助计划基金资助项目

MAXGDDP:decision data release with differential privacy

Jibin FU¹,Xiaojian ZHANG¹,Liping DING²

¹ College of Computer ＆Information Engineering,Henan University of Economics and Law,Zhengzhou 450046,China
² Institute of Software,Chinese Academy of Sciences,Beijing 100190,China

Revised:2018-02-27 Online:2018-03-01 Published:2018-04-02
Supported by:
The National Natural Science Foundation of China(61502146);The National Natural Science Foundation of China(91646203);The National Natural Science Foundation of China(91746115);The Natural Science Foundation of Henan Province(162300410006);The Key Technologies R＆D Program of Henan Province(142102210384);The Key Technologies R＆D Program of Henan Province(172102310713);The Research Program of The Higher Education of Henan Educational Committee(16A520002);Foundation for The Excellent Youth Teacher of Henan Province;The Young Talents Fund of Henan University of Economics and Law

摘要/Abstract

摘要：

基于层次细化的差分隐私决策数据发布得到了研究者的广泛关注，层次节点的选择、分类树的构建以及每层隐私代价的分配直接制约着决策数据发布结果的好坏，也影响最终的数据分析结果。针对现有基于层次细化的决策数据发布方法难以兼顾上述问题的不足，提出一种高效的分层细化方法MAXGDDP，该方法对原始分类数据进行分层细化，在同一层次的概念细化中提出了最大值属性索引算法，在不同层次之间利用类几何分配机制来更加合理地分配隐私预算。基于真实数据集对比了 MAXGDDP 与 DiffGen 算法，实验结果表明该方法在保护数据隐私的同时，提高了发布数据的分类准确率。

关键词: 决策数据, 数据发布, 差分隐私, 层次细化

Abstract:

Specialization-based private decision data release has attracted considerable research attention in recent years.The relation among hierarchical node,taxonomy tree,and budget allocation directly constrains the accuracy of data release and classification.Most existing methods based on hierarchical specialization cannot efficiently address the above problems.An effective method was proposed,called MAXGDDP to publish decision data with specialization.MAXGDDP employed MAX index attribute selection algorithm to select the highlight concept for furthering specialization in each hierarchy.Besides,for making more rational use of privacy budget,MAXGDDP relied on geometric strategy to allocate the privacy budget in each hierarchy.Compared with existing methods such as DiffGen on the real datasets,MAXGDDP outperforms its competitors,achieves data privacy and the better result of classification simultaneously.

Key words: decision data, data release, differential privacy, hierarchical specialization

中图分类号:

TP309

傅继彬,张啸剑,丁丽萍. MAXGDDP：基于差分隐私的决策数据发布算法[J]. 通信学报, 2018, 39(3): 136-146.

Jibin FU,Xiaojian ZHANG,Liping DING. MAXGDDP:decision data release with differential privacy[J]. Journal on Communications, 2018, 39(3): 136-146.

图/表 8

图1

图2

表1

表2

图3

图4

图5

图6

参考文献 21

[1]	SWEENEY L . k-anonymity:a model for protecting privacy[J]. International Journal of Uncertainty,Fuzziness and Knowledge-Based Systems, 2002,10(5): 557-570.
[2]	MACHANAVAJJHALA A , KIFER D , GEHRKE J ,et al. l-diversity:privacy beyond k-anonymity[J]. ACM Transactions on Knowledge Discovery from Data (TKDD), 2007,1(1): 3-52.
[3]	LI N , LI T , VENKATASUBRAMANIAN S . t-closeness:privacy beyond k-anonymity and l-diversity[C]// 2007 IEEE 23rd International Conference on Data Engineering. 2007: 106-115.
[4]	TERROVITIS M , MAMOULIS N , KALNIS P . Privacy-preserving anonymization of set-valued data[C]// Very Large Data Base Endowment. 2008: 115-125.
[5]	DWORK C , . Differential privacy[C]// 33rd International Colloquium on Automata,Languages and Programming,part II (ICALP 2006). 2006: 1-12.
[6]	DWORK C , LEI J . Differential privacy and robust statistics[C]// The 41th Annual ACM Symposium on Theory of Computing (STOC). 2009: 371-380.
[7]	FRIEDMAN A , SCHUSTER A . Data mining with differential privacy[C]// The 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2010: 493-502.
[8]	MOHAMMED N , CHEN R , FUNG B ,et al. Differentially private data release for data mining[C]// The 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2011: 493-501.
[9]	ZHU T , XIONG P , XIANG Y ,et al. An effective deferentially private data releasing algorithm for decision tree[C]// 12th IEEE International Conference on Trust,Security and Privacy in Computing and Communications. 2013: 388-395.
[10]	GANTA S R , KASIVISWANATHAN S P , SMITH A . Composition attacks and auxiliary information in data privacy[C]// The 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2008: 265-273.
[11]	ZHU T , LI G , ZHOU W ,et al. Differentially private data publishing and analysis:a survey[J]. IEEE Transactions on Knowledge and Data Engineering, 2017,29(8): 1619-1638.
[12]	熊平, 朱天清, 王晓峰 . 差分隐私保护及其应用[J]. 计算机学报, 2014,37(1): 101-122.
	XIONG P , ZHU T Q , WANG X F . A survey on differential privacy and applications[J]. Chinese Journal of Computers, 2014,37(1): 101-122.
[13]	康海燕, 马跃雷 . 差分隐私保护在数据挖掘中应用综述[J]. 山东大学学报:理学版, 2017(3): 16-23.
	KANG H Y , MA Y L . Survey on application of data ming via differential privacy[J]. Journal of Shangdong University(Natural Science), 2017(3): 16-23.
[14]	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.
	ZHANG X J , MENG X F . Differential privacy in data publication and analysis[J]. Chinese Journal of Computers, 2014,37(4): 927-949.
[15]	BLUM A , DWORK C , MCSHERRY F ,et al. Practical privacy:the SuLQ framework[C]// The Twenty-Fourth ACM SIGMOD-SIGACTSIGART Symposium on Principles of Database Systems. 2005: 128-138.
[16]	MCSHERRY F D , . Privacy integrated queries:an extensible platform for privacy-preserving data analysis[C]// The 2009 ACM SIGMOD International Conference on Management of Data. 2009: 19-30.
[17]	GOSWAMI P , MADAN S . Privacy preserving data publishing and data anonymization approaches:a review[C]// 2017 International Conference on Computing,Communication and Automation. 2017: 139-142.
[18]	FLETCHER S , ISLAM M Z . A differentially private random decision forest using reliable signal-to-noise ratios[C]// Australasian Joint Conference on Artificial Intelligence. 2015: 192-203.
[19]	CORMODE G , PROCOPIUC C , SRIVASTAVA D ,et al. Differentially private spatial decompositions[C]// IEEE 28th International Conference on Data Engineering. 2012: 20-31.
[20]	DWORK C , MCSHERRY F , NISSIM K ,et al. Calibrating noise to sensitivity in private data analysis[C]// Theory of Cryptography Conference. 2006: 265-284.
[21]	MCSHERRY F , TALWAR K . Mechanism design via differential privacy[C]// 48th Annual IEEE Symposium on Foundations of Computer Science. 2007: 94-103.

属性	类型
sepal length	数值型
sepal width	数值型
petal length	数值型
petal width	数值型

属性	类型
age	数值型
workclasst	非数值型
fnlwgt	数值型
education	非数值型
education-num	数值型
martial-status	非数值型
occupation	非数值型
relationship	非数值型
race	非数值型
sex	非数值型
capital-gain	数值型
capital-loss	数值型
hours-per-week	数值型
Native-country	非数值型

MAXGDDP：基于差分隐私的决策数据发布算法

MAXGDDP:decision data release with differential privacy

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 21

相关文章 15

Metrics

推荐阅读 0

[1]	余晟兴, 陈泽凯, 陈钟, 刘西蒙. DAGUARD：联邦学习下的分布式后门攻击防御方案[J]. 通信学报, 2023, 44(5): 110-122.
[2]	冯涛, 陈李秋, 方君丽, 石建明. 基于本地化差分隐私和属性基可搜索加密的区块链数据共享方案[J]. 通信学报, 2023, 44(5): 224-233.
[3]	张淑芬, 董燕灵, 徐精诚, 王豪石. 基于目标扰动的AdaBoost算法[J]. 通信学报, 2023, 44(2): 198-209.
[4]	汤凌韬, 王迪, 刘盛云. 面向非独立同分布数据的联邦学习数据增强方案[J]. 通信学报, 2023, 44(1): 164-176.
[5]	袁程胜, 郭强, 付章杰. 基于差分隐私的深度伪造指纹检测模型版权保护算法[J]. 通信学报, 2022, 43(9): 181-193.
[6]	王瀚仪, 李效光, 毕文卿, 陈亚虹, 李凤华, 牛犇. 多级本地化差分隐私算法推荐框架[J]. 通信学报, 2022, 43(8): 52-64.
[7]	张勇, 李丹丹, 韩璐, 黄小红. 隐私保护的群体感知数据交易算法[J]. 通信学报, 2022, 43(5): 1-13.
[8]	康海燕, 冀源蕊. 基于本地化差分隐私的联邦学习方法研究[J]. 通信学报, 2022, 43(10): 94-105.
[9]	晏燕, 丛一鸣, Adnan Mahmood, 盛权政. 基于深度学习的位置大数据统计发布与隐私保护方法[J]. 通信学报, 2022, 43(1): 203-216.
[10]	陈思, 付安民, 苏铓, 孙怀江. 基于差分隐私的轨迹隐私保护方案[J]. 通信学报, 2021, 42(9): 54-64.
[11]	李洪涛, 任晓宇, 王洁, 马建峰. 基于差分隐私的连续位置隐私保护机制[J]. 通信学报, 2021, 42(8): 164-175.
[12]	蔡剑平, 刘西蒙, 熊金波, 应作斌, 吴英杰. 差分隐私下多重一致性约束问题的逼近方法[J]. 通信学报, 2021, 42(6): 107-117.
[13]	朱素霞, 刘抒伦, 孙广路. 基于相对熵和K-means的形状相似差分隐私轨迹保护机制[J]. 通信学报, 2021, 42(2): 113-123.
[14]	刘杨, 李珺, 陈文韵, 彭木根. 面向6G的雾无线接入网内生安全数据共享机制研究[J]. 通信学报, 2021, 42(1): 67-78.
[15]	叶阿勇,孟玲玉,赵子文,刁一晴,张娇美. 基于预测和滑动窗口的轨迹差分隐私保护机制[J]. 通信学报, 2020, 41(4): 123-133.