计算机工程 ›› 2019, Vol. 45 ›› Issue (11): 126-132.doi: 10.19678/j.issn.1000-3428.0052660

• 安全技术 • 上一篇    下一篇

基于关联规则的多敏感属性匿名算法

吴睿雪a,b,c, 彭长根a,b,c, 刘波涛a,b,c, 丁红发b,c,d, 谢明明b,c,d   

  1. 贵州大学 a. 计算机科学与技术学院;b. 贵州省公共大数据重点实验室;c. 密码学与数据安全研究所;d. 数学与统计学院, 贵阳 550025
  • 收稿日期:2018-09-14 修回日期:2018-10-22 发布日期:2018-11-01
  • 作者简介:吴睿雪(1995-),女,硕士研究生,主研方向为隐私保护、数据安全;彭长根(通信作者),教授、博士、博师生导师;刘波涛,硕士研究生;丁红发,讲师、博士研究生;谢明明,硕士研究生。
  • 基金项目:
    国家自然科学基金(61662009,61772008);"十三五"国家密码发展基金(MMJJ20170129);贵州省科技计划项目(黔科合基础[2016]2315,黔科合基础[2017]1045);贵州省科技计划项目(黔科合重大专项[2017]3002,黔科合重大专项[2018]3001)。

Multi-sensitive Attribute Anonymity Algorithm Based on Association Rule

WU Ruixuea,b,c, PENG Changgena,b,c, LIU Botaoa,b,c, DING Hongfab,c,d, XIE Mingmingb,c,d   

  1. a. College of Computer Science and Technology;b. Guizhou Provincial Key Laboratory of Big Data;c. Institute of Cryptography and Data Security;d. College of Mathematics and Statistics, Guizhou University, Guiyang 550025, China
  • Received:2018-09-14 Revised:2018-10-22 Published:2018-11-01

摘要: 针对多数隐私保护算法不能较好平衡数据精度和数据隐私保护程度的问题,从数据集中准标识属性与敏感属性的关联关系出发,提出一种基于关联规则的匿名算法。运用Aprior算法建立属性间的关联规则,利用互信息量度量其关联度,为准标识属性的分级分类提供依据,同时设置泛化边界与权重,以避免产生较大的匿名成本。实验结果表明,该算法能够减少数据损失,实现数据效用与隐私保护之间的均衡。

关键词: 隐私保护, 多敏感属性, 关联关系, 泛化边界, 关联规则

Abstract: To address the problem that most privacy protection algorithms can not balance the data accuracy and data privacy protection degree,an anonymity algorithm based on association rules is proposed according to the association relation between quasi-identification attributes and sensitive attributes in data sets.The Aprior algorithm is used to establish the association rules between attributes,and the mutual information is used to measure the degree of association,which provides a basis for the classification of quasi-identification attributes.Also generalized boundaries and weights are set to avoid large anonymity costs.Experimental results show that the algorithm can reduce data loss and achieve the balancing between data utility and privacy protection.

Key words: privacy protection, multi-sensitive attribute, association relation, generalized boundary, association rule

中图分类号: