摘要
为了在保护隐私的前提下提高发布数据的质量,该文在基于空间划分的隐私保护数据发布技术基础上,提出一种基于空间混合划分技术的隐私保护关系型数据发布方法。通过对传统基于空间严格划分与非严格划分技术的关系型数据匿名发布问题的进一步研究与分析发现,严格划分技术所导致的信息损失高于非严格划分技术,而非严格划分技术所发布的数据可能存在查询混淆,从而在数据可用性上可能不如严格划分技术。该文结合严格划分与非严格划分的优点,提出一种基于混合划分技术的隐私保护关系型数据发布算法。实验对该文算法所发布数据的质量及算法效率与同类算法进行比较分析。实验结果表明,该文算法是有效可行的。
To increase the released data quality while assuring the privacy security,this paper proposes a space hybrid partitioning approach for privacy preserving relational data publication.After further research and analysis on privacy preserving relational data publication based on strict and non-strict space partitioning.This paper finds that the information loss of the released data produced by strict partitioning is higher than the information loss of the released data produced by non-strict partitioning,while some query confusion may occur in the released data by non-strict partitioning,thus leading to higher data quality of the release data produced by strict partitioning than by non-strict partitioning.In this paper,a hybrid partitioning approach for privacy preserving relational data publication,which combines the advantages of strict partitioning and non-strict partitioning,is presented.Experimental analysis is designed by comparing the algorithm proposed here and the traditional algorithms on the released data availability and the algorithm efficiency.Experimental results show that the proposed algorithm is effective and feasible.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2013年第4期493-499,共7页
Journal of Nanjing University of Science and Technology
基金
福建省自然科学基金(2010J01330)
福州大学科技发展基金(2012-XQ-27)
关键词
隐私保护
关系型数据发布
k匿名算法
混合划分
数据质量
privacy preservation
relational data publication
anonymous algorithm
hybrid partitioning
data quality
