面向多元组关系-集值数据的脱敏方法  

Data masking approach for multiple tuples with relational and set-valued attributes

作　　者：顾兆军[1] 蔡畅 刘春波 钟安鸣[2] GU Zhao-jun;CAI Chang;LIU Chun-bo;ZHONG An-ming(Information Security Evaluation Center,Civil Aviation University of China,Tianjin 300300,China;College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学信息安全测评中心,天津300300 [2]中国民航大学计算机科学与技术学院,天津300300

出　　处：《计算机工程与设计》2022年第1期34-42,共9页Computer Engineering and Design

基　　金：国家自然科学基金项目(61601467、U1533104);民航安全能力建设基金项目(PE2017003、AADSA2019026);中央高校基本科研业务费中国民航大学专项基金项目(3122014D033)。

摘　　要：对数据发布中传统方法脱敏多元组关系-集值数据可能导致信息泄露以及产生较高信息损失的问题进行研究,提出基于(K,L)-多样性模型的多元组关系-集值数据的脱敏方法PAHI。根据准标识符将多元组数据转换为单元组数据;用信息增益比优化分割方法,实现集值数据K-匿名;引入敏感度值建立集值指纹桶,采用敏感度距离优化剩余元组的处理,实现关系-集值数据L-多样性。引入辨识度度量标准和信息熵计算K和L的阈值。安全性分析和实验验证,对比现有的脱敏方法,该方法保证了脱敏后数据安全,降低了时间开销和信息损失,提高了数据的可用性。Aiming at the problem that the traditional method of masking multiple tuples with relational and set-valued attributes may lead to information leakage and high information loss in the data release,a data masking approach for multiple tuples with relational and set-valued attributes based on(K,L)-diversity which was called PAHI was proposed.According to the quasi-identifiers,the multiple tuples data were transformed into the unit tuple data.The K-anonymity of data with set-valued attribute was realized by optimizing the partition method of information gain ratio.The sensitivity value was introduced to establish the set-valued fingerprint bucket,and the sensitivity distance was used to optimize the processing of the remaining tuples to realize the L-diversity of data with relational and set-valued attributes.The threshold values of K and L were calculated by using the discernability metric and information entropy.Safety analysis and experimental results show that the proposed method ensures the data security after masking,reduces the time cost and information loss,and improves the data availability,compared with the existing desensitization methods.

关 键 词：数据脱敏 匿名化 (K L)-多样性 关系-集值数据 多元组数据 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]