基于概率统计与德尔熵值法的隐私保护综合评价模型
|
摘要
云计算环境中,租户将数据存储于SaaS(Software as a Service)应用平台中,利用分块混淆的隐私保护技术将数据切分为多个数据块并存储到不同的存储模式中,实现明文状态下租户数据的分离与保护.但是这种隐私保护技术的隐私保护程度如何,用户是无法明确感知的,为此,针对分块混淆的隐私保护技术,该文提出一种基于概率统计与德尔熵值法的隐私保护综合评价模型.首先分析了基于分块混淆的隐私保护技术的当前状况,在此基础上定义了隐私保护评价指标,利用概率统计的知识,定义评价指标的计算规则,构建隐私保护层次分析模型,通过由数据块层到数据存储模式层(Data Storage Mode,DSM)层再到顶层回逆的方式,得到隐私保护后数据分布的评价指标值;然后分析了德尔菲法和熵值法在权重确定方面的优势,将德尔菲法的主观判别与熵值法的客观判断相结合,改进两种方法的计算过程,提出基于德尔熵值法的指标权重确定模型,得到隐私保护效果评价指标的权重;最后定义评价等级,建立基于概率统计和德尔熵值法的隐私保护综合评价模型,实现基于分块混淆的隐私保护技术的综合评价.实验结果证明本文提出的综合评价模型不仅可以客观地评价基于分块混淆的隐私保护技术的隐私保护效果,也证明了分块混淆隐私保护技术的有效性,为SaaS应用平台中的数据隐私保护提供了强有力的理论支撑.
In the environment of cloud computing,the data is stored on the SaaS application platform by users.On the cloud-based SaaS application support platform,using block confusing privacy protection technology,data is divided into multiple data blocks and stored in different storage patterns,which achieves the separation and protection of privacy data.At the same time,the correlation of the data slices is stored on the reliable third party's platform by introducing trusted third party,which achieves the separation and protecting of private data in clear text.However,users are not clear to what degree their private information is protected by using this privacy protection method.In order to solve this confusion,this paper proposes a comprehensive privacy preservation evaluation model to present an intuitive impression on block confusing privacy protection technology to users based on probability and Del entropy method.First of all,this paper defines the evaluation index of privacy protection by analyzing the current status of privacy protection technologies.Based on the knowledge of probability and statistics,the calculation rules of evaluation index are defined and the model of privacy protection levels composed of data blocks,Data Storage Mode(DSM),and privacy protection is constructed.By the method of backtracking and calculating the privacy protection evaluation index of each data blocks,the DSM layer privacy protection evaluation index,and the top privacy protection index,the evaluation index value of the data distribution after each tenant adopting the privacy protection policy is obtained.Then,using the weighting advantages of Delphi method and entropy method,the subjective judgments of Delphi method and the objective judgments of entropy method is combined.On the basis of improving the calculation process of these two methods,this paper proposes an index weight determination model based on Del-entropy method.By this model,the weights of the privacy protection effectiveness evaluation indicators are obtained.At last,the levels of evaluation are defined,the privacy preservation comprehensive evaluation model of privacy protection based on probabilistic statistics and Del-entropy method is set up to realize the intuitionistic evaluation of blocking confusion privacy protection methods,and a single-factor evaluation algorithm,evaluation index weight distribution algorithm,and privacy protection fuzzy comprehensive evaluation algorithm are proposed to calculate the privacy protection level of tenant data.In order to verify the correctness of the privacy protection evaluation index proposed in this paper and the effectiveness of the privacy protection evaluation model based on probability statistics and Del-entropy method,experiments are performed by using the privacy protection calculation method.The experimental results show that the comprehensive evaluation model proposed in this paper can not only objectively demonstrate the privacy protection effects of privacy protection technology based on block obfuscation,but also prove the effectiveness of blocking confidential privacy protection methods and pointed out the direction of improvement.The most important is that it provides a good theory for data privacy protection in SaaS application platform.
In the environment of cloud computing,the data is stored on the SaaS application platform by users.On the cloud-based SaaS application support platform,using block confusing privacy protection technology,data is divided into multiple data blocks and stored in different storage patterns,which achieves the separation and protection of privacy data.At the same time,the correlation of the data slices is stored on the reliable third party's platform by introducing trusted third party,which achieves the separation and protecting of private data in clear text.However,users are not clear to what degree their private information is protected by using this privacy protection method.In order to solve this confusion,this paper proposes a comprehensive privacy preservation evaluation model to present an intuitive impression on block confusing privacy protection technology to users based on probability and Del entropy method.First of all,this paper defines the evaluation index of privacy protection by analyzing the current status of privacy protection technologies.Based on the knowledge of probability and statistics,the calculation rules of evaluation index are defined and the model of privacy protection levels composed of data blocks,Data Storage Mode(DSM),and privacy protection is constructed.By the method of backtracking and calculating the privacy protection evaluation index of each data blocks,the DSM layer privacy protection evaluation index,and the top privacy protection index,the evaluation index value of the data distribution after each tenant adopting the privacy protection policy is obtained.Then,using the weighting advantages of Delphi method and entropy method,the subjective judgments of Delphi method and the objective judgments of entropy method is combined.On the basis of improving the calculation process of these two methods,this paper proposes an index weight determination model based on Del-entropy method.By this model,the weights of the privacy protection effectiveness evaluation indicators are obtained.At last,the levels of evaluation are defined,the privacy preservation comprehensive evaluation model of privacy protection based on probabilistic statistics and Del-entropy method is set up to realize the intuitionistic evaluation of blocking confusion privacy protection methods,and a single-factor evaluation algorithm,evaluation index weight distribution algorithm,and privacy protection fuzzy comprehensive evaluation algorithm are proposed to calculate the privacy protection level of tenant data.In order to verify the correctness of the privacy protection evaluation index proposed in this paper and the effectiveness of the privacy protection evaluation model based on probability statistics and Del-entropy method,experiments are performed by using the privacy protection calculation method.The experimental results show that the comprehensive evaluation model proposed in this paper can not only objectively demonstrate the privacy protection effects of privacy protection technology based on block obfuscation,but also prove the effectiveness of blocking confidential privacy protection methods and pointed out the direction of improvement.The most important is that it provides a good theory for data privacy protection in SaaS application platform.
引文
[1]Zhang Kun,Li Qing-Zhong,Shi Yu-Liang.Research on data combination privacy preservation mechanism for SaaS.Chinese Journal of Computers,2010,33(11):2044-2054(in Chinese)(张坤,李庆忠,史玉良.面向SaaS应用的数据组合隐私保护机制研究.计算机学报,2010,33(11):2044-2054)
[2]Shao Ya-Li,Shi Yu-Liang,Li Hui.A novel cloud data fragmentation cluster-based privacy preserving mechanism.International Journal of Grid&Distributed Computing,2014,7(4):21-32
[3]Shi Yu-Liang,Jiang Zhen,Zhang Kun.Policy-Based customized privacy preserving mechanism for SaaS applications//Proceedings of the International Conference on Grid and Pervasive Computing.Berlin,Germany,2013:491-500
[4]Khan K M,Shaheen M.Data obfuscation for privacy and confidentiality in cloud computing//Proceedings of the IEEEInternational Conference on Software Quality,Reliability and Security-Companion.Vancouver,Canada,2015:195-196
[5]Wang Yu-Li,Tian Jiay-Jin,Yang Cheng,Zhu Ya-Ping.Research on anonymous protection technology for big data publishing//Proceedings of the 7th IEEE International Conference on Software Engineering and Service Science.Beijing,China,2016:438-441
[6]Shalin Eliabeth S,Sarju S.Bigdata anonymization using one dimensional and multidimensional map reduce framework on cloud.International Journal of Database Theory and Application,2015,8(6):253-262
[7]Tran Q,Sato H.A solution for privacy protection in mapreduce//Proceedings of the 36th IEEE Annual International Computer Software and Applications Conference.Izmir,Turkey,2012:515-520
[8]ádám Erdélyi,Winkler T,Rinner B.Privacy protection vs.utility in visual data:An objective evaluation framework.Multimedia Tools&Applications,2018,77(2):2285-2312
[9]Shi Yu-Liang,Chen Yu,Zhou Zhong-Min,Cui Li-Zhen.Aprivacy protection evaluation mechanism for dynamic data based on chunk-confusion.Journal of Signal Processing Systems,2017,89(1):27-39
[10]Tahir N,James F.An annotation-free method for evaluating privacy protection techniques in videos//Proceedings of the IEEE Computer Society 2015.ISBN,Karlsruhe,Germany,2015:1-6
[11]Hugues D L S-G,David D,Guillaume R.Range reduction based on Pythagorean triples for trigonometric function evaluation//Proceedings of the 26th IEEE International Conference on Application-Specific Systems,Architectures and Processors.Toronto,Canada,2015:74-81
[12]Wang Yu-Biao,Wen Jun-Hao,Wang Xi-Bin,Zhou Wei.Cloud service evaluation model based on trust and privacyaware.Optik-International Journal for Light and Electron Optics,2017,134:269-279
[13]Zhu Jian-Ming,Srinivasan R.Evaluation model of information security technologies based on game theoretic.Chinese Journal of Computers,2009,32(4):828-834(in Chinese)(朱建明,Srinivasan R.基于博弈论的信息安全技术评价模型.计算机学报,2009,32(4):828-834)
[14]Jiao Yu-Min,Wang Qiang,Su Fan-Tun,et al.Research on dynamic evaluation method for process-oriented virtual training.Acta Armamentarii,2012,33(7):875-880
[15]Lund J R,Israel M.Information theory and multi-objective evaluation//Proceedings of the 1992National Conference on Water Resources Planning and Management-Water Forum’92.Baltimore,USA,1992:486-491
[16]Hamed A,Ben Ayed H K.Privacy risk assessment for Web tracking:A user-oriented approach toward privacy risk assessment for Web tracking//Proceedings of the 2016IEEECanadian Conference on Electrical and Computer Engineering.Vancouver,Canada,2016:1-6
[17]Ren Dan-Dan,Du Su-Guo,Zhu Hao-Jin.A novel attack tree based risk assessment approach for location privacy preservation in the VANETs//Proceedings of the 2011IEEE International Conference on Communications.Kyoto,Japan,2011:1-5
[18]Chen Yan-Bo,Liu Yang,Zhang Ji,et al.An evaluation method of design scheme for new generation smart substation based on improved SEC.Power System Technology,2017,41(4):1308-1314
[19]Xue Zhe-Bin,Zeng Xian-Yi,Koehl L.Development of a method based on fuzzy comprehensive evaluation and genetic algorithm to study relations between tactile properties and total preference of textile products.Journal of the Textile Institute,2016,108(7):1085-1094
[20]Wang Shao-Na,Dong Rui,Xie Hui,et al.The application of Delphi method and its construction index system.Journal of Bengbu Medical College,2016,41(5):695-698(in Chinese)(王少娜,董瑞,谢晖等.德尔菲法及其构建指标体系的应用进展.蚌埠医学院学报,2016,41(5):695-698)
[21]Wang Sheng-Chang,Fu Di,Chen Juan-Juan,et al.A method to determine subjective evaluation index of vehicle dynamic performance based on entropy method.Journal of Highway and Transportation Research and Development,2015,32(7):153-158(in Chinese)(王生昌,付迪,陈娟娟等.基于熵值法的汽车动力性能主观评价指标权重确定方法.公路交通科技,2015,32(7):153-158)
[22]Zhang Kun,Abraham A,Shi Yu-Liang.Data combination privacy preservation adjusting mechanism for software as a service//Proceedings of the IEEE International Conference on Systems,Man,and Cybernetics.Manchester,UK,2013:2007-2012
[23]Chen Yu,Shi Yu-Liang,Cui Li-Zhen.Model for hiding data relationships based on Chunk-Confusion in cloud computing//Proceedings of the 2016IEEE Trustcom/BigDataSE/ISPA.Tianjin,China,2016:837-844
(1)http://www.baike.com/wiki/全概率公式.
(1)http://wiki.mbalib.com/wiki/德尔菲法.
(2)http://wiki.mbalib.com/wiki/熵值法.
(3)http://wiki.mbalib.com/wiki/决策矩阵.
[2]Shao Ya-Li,Shi Yu-Liang,Li Hui.A novel cloud data fragmentation cluster-based privacy preserving mechanism.International Journal of Grid&Distributed Computing,2014,7(4):21-32
[3]Shi Yu-Liang,Jiang Zhen,Zhang Kun.Policy-Based customized privacy preserving mechanism for SaaS applications//Proceedings of the International Conference on Grid and Pervasive Computing.Berlin,Germany,2013:491-500
[4]Khan K M,Shaheen M.Data obfuscation for privacy and confidentiality in cloud computing//Proceedings of the IEEEInternational Conference on Software Quality,Reliability and Security-Companion.Vancouver,Canada,2015:195-196
[5]Wang Yu-Li,Tian Jiay-Jin,Yang Cheng,Zhu Ya-Ping.Research on anonymous protection technology for big data publishing//Proceedings of the 7th IEEE International Conference on Software Engineering and Service Science.Beijing,China,2016:438-441
[6]Shalin Eliabeth S,Sarju S.Bigdata anonymization using one dimensional and multidimensional map reduce framework on cloud.International Journal of Database Theory and Application,2015,8(6):253-262
[7]Tran Q,Sato H.A solution for privacy protection in mapreduce//Proceedings of the 36th IEEE Annual International Computer Software and Applications Conference.Izmir,Turkey,2012:515-520
[8]ádám Erdélyi,Winkler T,Rinner B.Privacy protection vs.utility in visual data:An objective evaluation framework.Multimedia Tools&Applications,2018,77(2):2285-2312
[9]Shi Yu-Liang,Chen Yu,Zhou Zhong-Min,Cui Li-Zhen.Aprivacy protection evaluation mechanism for dynamic data based on chunk-confusion.Journal of Signal Processing Systems,2017,89(1):27-39
[10]Tahir N,James F.An annotation-free method for evaluating privacy protection techniques in videos//Proceedings of the IEEE Computer Society 2015.ISBN,Karlsruhe,Germany,2015:1-6
[11]Hugues D L S-G,David D,Guillaume R.Range reduction based on Pythagorean triples for trigonometric function evaluation//Proceedings of the 26th IEEE International Conference on Application-Specific Systems,Architectures and Processors.Toronto,Canada,2015:74-81
[12]Wang Yu-Biao,Wen Jun-Hao,Wang Xi-Bin,Zhou Wei.Cloud service evaluation model based on trust and privacyaware.Optik-International Journal for Light and Electron Optics,2017,134:269-279
[13]Zhu Jian-Ming,Srinivasan R.Evaluation model of information security technologies based on game theoretic.Chinese Journal of Computers,2009,32(4):828-834(in Chinese)(朱建明,Srinivasan R.基于博弈论的信息安全技术评价模型.计算机学报,2009,32(4):828-834)
[14]Jiao Yu-Min,Wang Qiang,Su Fan-Tun,et al.Research on dynamic evaluation method for process-oriented virtual training.Acta Armamentarii,2012,33(7):875-880
[15]Lund J R,Israel M.Information theory and multi-objective evaluation//Proceedings of the 1992National Conference on Water Resources Planning and Management-Water Forum’92.Baltimore,USA,1992:486-491
[16]Hamed A,Ben Ayed H K.Privacy risk assessment for Web tracking:A user-oriented approach toward privacy risk assessment for Web tracking//Proceedings of the 2016IEEECanadian Conference on Electrical and Computer Engineering.Vancouver,Canada,2016:1-6
[17]Ren Dan-Dan,Du Su-Guo,Zhu Hao-Jin.A novel attack tree based risk assessment approach for location privacy preservation in the VANETs//Proceedings of the 2011IEEE International Conference on Communications.Kyoto,Japan,2011:1-5
[18]Chen Yan-Bo,Liu Yang,Zhang Ji,et al.An evaluation method of design scheme for new generation smart substation based on improved SEC.Power System Technology,2017,41(4):1308-1314
[19]Xue Zhe-Bin,Zeng Xian-Yi,Koehl L.Development of a method based on fuzzy comprehensive evaluation and genetic algorithm to study relations between tactile properties and total preference of textile products.Journal of the Textile Institute,2016,108(7):1085-1094
[20]Wang Shao-Na,Dong Rui,Xie Hui,et al.The application of Delphi method and its construction index system.Journal of Bengbu Medical College,2016,41(5):695-698(in Chinese)(王少娜,董瑞,谢晖等.德尔菲法及其构建指标体系的应用进展.蚌埠医学院学报,2016,41(5):695-698)
[21]Wang Sheng-Chang,Fu Di,Chen Juan-Juan,et al.A method to determine subjective evaluation index of vehicle dynamic performance based on entropy method.Journal of Highway and Transportation Research and Development,2015,32(7):153-158(in Chinese)(王生昌,付迪,陈娟娟等.基于熵值法的汽车动力性能主观评价指标权重确定方法.公路交通科技,2015,32(7):153-158)
[22]Zhang Kun,Abraham A,Shi Yu-Liang.Data combination privacy preservation adjusting mechanism for software as a service//Proceedings of the IEEE International Conference on Systems,Man,and Cybernetics.Manchester,UK,2013:2007-2012
[23]Chen Yu,Shi Yu-Liang,Cui Li-Zhen.Model for hiding data relationships based on Chunk-Confusion in cloud computing//Proceedings of the 2016IEEE Trustcom/BigDataSE/ISPA.Tianjin,China,2016:837-844
(1)http://www.baike.com/wiki/全概率公式.
(1)http://wiki.mbalib.com/wiki/德尔菲法.
(2)http://wiki.mbalib.com/wiki/熵值法.
(3)http://wiki.mbalib.com/wiki/决策矩阵.