Data collection method based on personalized local differential privacy

Abstract

The invention belongs to the field of information security, and particularly relates to a data collection method based on personalized local differential privacy. The method comprises the steps that a server divides an original data set and sets a plurality of privacy levels, and the division result and the privacy levels are open. A user selects a privacy level, performs coding disturbance on original data of the user locally, and sends the disturbed data to the server. The server collects the data and performs statistical analysis under each privacy level, estimates a frequency distribution result of the original data, and performs weighted combination on the results under each privacy level to obtain a final frequency distribution result. According to the method, attackers with any background knowledge can be resisted, and privacy attacks from an untrusted third party can be prevented; and according to the method, personalization is achieved from the user level, the sensitivity difference of different data is considered, the user can control the privacy protection strength by himself / herself, the data cannot be over-protected, and the estimation result obtained by the server is more accurate.

Description

technical field [0001] The invention belongs to the field of information security, and in particular relates to a data collection method based on personalized local differential privacy. Background technique [0002] With the rapid development of information technology, more and more personal information is collected and analyzed for various purposes. For example, people's location information will be collected for route planning or scenic spot recommendation; people's medical records will be collected for health risk assessment or expected disease diagnosis. Although these behaviors bring great convenience to users, they also cause privacy leakage problems, such as leaking personal home addresses or health information. In this environment, how to protect personal information has become a common concern of the society. [0003] Differential privacy is a privacy protection model proposed by Dwork [DWORK, C.Differential privacy.In ICALP (2006), pp.1-12.], which is different ...

AI Technical Summary

Problems solved by technology

[0005] However, most of the existing local differential privacy mechanisms have two problems when they are used: one is that they do not take into account the user's individual privacy protection needs

This approach ignores the differences in sensitivity between different types of data, and perturbs them in the same perturbation method, which will lead to over-protection of low-sensitivity data and reduce the accuracy of the final statistical results; or high-sensitivity data Privacy protection needs are not met, increasing the risk of privacy leakage

Images