The Privacy-Utility Tradeoff in Rank-Preserving Dataset Obfuscation

The Privacy-Utility Tradeoff in Rank-Preserving Dataset Obfuscation
复制标题

DOI:
10.1109/isit54713.2023.10206447
复制
发表时间:
2023-05
期刊:
2023 IEEE International Symposium on Information Theory (ISIT)
影响因子:
--
通讯作者:
Mahshad Shariatnasab;Farhad Shirani;S. S. Iyengar-S.
Mahshad Shariatnasab;Farhad Shirani;S. S. Iyengar-S.
中科院分区:
其他
文献类型:
--
作者:
Mahshad Shariatnasab;Farhad Shirani;S. S. Iyengar-S.

文献摘要

相似文献

数据集混淆是指在公开发布之前将随机噪声添加到给定数据集的条目中以防止私人信息泄露的技术。在这项工作中,数据集混淆下的两个目标被认为是:i)排名保持:以保持由给定的排名函数引起的混淆数据集中的行排序,和ii)匿名性:以保护用户匿名指纹攻击。第一个目标,排名保护,是感兴趣的应用程序,如搜索引擎和推荐系统的设计,功能匹配,和社会网络分析。在评估匿名目标时考虑的指纹攻击是隐私攻击,其中攻击者基于其观察到的活动(例如在线Web活动)构建受害者的指纹,并将此指纹与从公开发布的混淆数据集中提取的信息进行比较以识别受害者。通过评估一类混淆机制在渐近大数据集上的性能极限,量化了等级保持和用户匿名之间的基本权衡。考虑单字母混淆机制,其中数据集中的每个条目都受到独立噪声的干扰,其基本性能限制的特征在于利用大偏差技术。最优混淆测试通道,优化隐私效用权衡,其特征在于在一个凸优化问题,可以有效地解决的形式。各种场景的数值模拟提供了验证的理论推导。
Dataset obfuscation refers to techniques in which random noise is added to the entries of a given dataset, prior to its public release, to protect against leakage of private information. In this work, dataset obfuscation under two objectives is considered: i) rank-preservation: to preserve the row ordering in the obfuscated dataset induced by a given rank function, and ii) anonymity: to protect user anonymity under fingerprinting attacks. The first objective, rank-preservation, is of interest in applications such as the design of search engines and recommendation systems, feature matching, and social network analysis. Fingerprinting attacks, considered in evaluating the anonymity objective, are privacy attacks where an attacker constructs a fingerprint of a victim based on its observed activities, such as online web activities, and compares this fingerprint with information extracted from a publicly released obfuscated dataset to identify the victim. By evaluating the performance limits of a class of obfuscation mechanisms over asymptotically large datasets, a fundamental trade-off is quantified between rank-preservation and user anonymity. Single-letter obfuscation mechanisms are considered, where each entry in the dataset is perturbed by independent noise, and their fundamental performance limits are characterized by leveraging large deviation techniques. The optimal obfuscating test-channel, optimizing the privacy-utility tradeoff, is characterized in the form of a convex optimization problem which can be solved efficiently. Numerical simulations of various scenarios are provided to verify the theoretical derivations.