一种联邦隐私保护用户信息匹配算法

A Federated Privacy-Preserving User Information Matching Algorithm

信息化时代,人们在生产和生活中产生了大量的数据,但大多分散在不同的机构或组织中,难以对其建立关联。现有的基于隐私保护可达性查询的用户信息匹配算法大多假设存在一个拥有完整数据集的数据提供方,但现实中数据往往分散在不同参与方中,需要联合所有参与方的数据才能构成完整数据集。为解决这个问题,本文提出了一种联邦隐私保护用户信息匹配算法。在人工数据集中的运行结果显示,本文的方法能在数据分散的场景下进行用户信息匹配,同时加入的本地提前筛选阶段能有效减少待验证自然人数量,减少通信开销。

With the advent of the information age,people produce vast amounts of data in their daily lives and work,yet the majority of this data is dispersed across numerous institutions or organizations,making it challenging to link them together.Most of the existing user information matching algorithms based on privacy-preserving reachability queries assume that there is a data provider with a complete data set,but in reality,the data is often scattered among different participants,and it is necessary to combine the data of all participants to form a complete data set.To solve this problem,this paper proposes a federated privacy protection user information matching algorithm.The operation results in a manual dataset show that the method in this article can match user information in a scenario where data is scattered,and the addition of a local early screening stage can effectively reduce the number of natural persons to be verified and reduce communication overhead.