摘要
现实应用中,许多数据需要不断更新并动态发布。使用这些数据资源进行数据挖掘,在提供便捷服务的同时也会导致个人隐私信息泄露。差分隐私保护技术能防止攻击者进行任意背景知识攻击并提供有效的隐私保护,因此成为数据发布领域中的研究热点。阐述了差分隐私保护技术的原理及性质;重点针对动态数据发布的差分隐私保护技术进行分类研究,并从关键技术、效果等方面对实时数据流发布和连续数据发布进行综合对比分析;总结了差分隐私保护在其他领域的应用并提出了未来的研究方向。
In many applications, many data needs to be continuously updated and dynamically released. However, using these data resources for data mining will lead to the disclosure of private information while providing convenient services. Differential privacy provides privacy protection with strong privacy guarantees against adversaries with any background knowledge. Therefore, it has become a hot topic in the field of data publishing. This paper first expounds the basic principles and characteristics of differential privacy protection technology, and then classifies the existing research results of differential privacy protection technology for dynamic data publication, focusing on dynamic data stream publication and continuous data publication. Finally, the application of differential privacy protection in other fields is summarized and the future research direction is proposed.
作者
屈晶晶
蔡英
夏红科
QU Jingjing;CAI Ying;XIA Hongke(Computer School,Beijing Information Science&Technology University,Beijing 100101,China)
出处
《北京信息科技大学学报（自然科学版）》
2019年第6期30-36,共7页
Journal of Beijing Information Science and Technology University
基金
国家自然科学基金资助项目(61672106)
北京市科技计划中央领导地方科技发展专项基金(Z171100004717002)
关键词
动态数据发布
差分隐私保护
动态数据流
连续数据发布
dynamic data publication
differential privacy
dynamic data stream
continuous data publication
