ON THE STATISTICAL SECURITY OF A STATISTICAL DATABASE.

被引：0

作者：

Kabe, D.G. ^{[1
]}

机构：

[1] St. Mary's Univ, Halifax, NS, Can, St. Mary's Univ, Halifax, NS, Can

来源：

Industrial Mathematics | 1987年 / 37卷 / pt 2期

关键词：

DATA PROCESSING - Security of Data - MATHEMATICAL TECHNIQUES - Perturbation Techniques - STATISTICAL METHODS;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

J. F. Traub et al. propose a statistical perturbation scheme to protect a statistical database against compromise. Their scheme handles the security of numerical as well as nonnumerical sensitive fields or a combination of fields. They use the usual Chebyshev inequality to analyze the trade-offs among the perturbations, the error incurred by statistical queries, and the query set to which they apply. By using a certain flexible form of Chebyshev's inequality, we show that the precision of the data information extracted through statistical queries can always be improved, thus ensuring more accurate results than the ones given by Traub et al.

引用

页码：119 / 129