Investigation method with confidence based on localized differential privacy

Abstract

The invention relates to an investigation method with confidence based on localized differential privacy, and belongs to the field of information security. According to the method, answers submitted by participants and corresponding confidence coefficients are protected by adopting a localized differential privacy method, attackers with any background knowledge can be resisted, and privacy attacksfrom untrusted third-party data collectors can be prevented; according to the invention, the confidence coefficient is utilized to quantify the confirmation degree of the participant to the answer submitted by the participant, and the accuracy of the statistical result is improved; according to the invention, after the participant selects one option and the corresponding confidence coefficient, the method of equally sharing the residual confidence coefficient is adopted, so that the expectation of the sum of statistical results of all answers is consistent with that of a traditional investigation scheme and is the total number of users participating in investigation.

Description

technical field [0001] The invention belongs to the field of information security, and in particular relates to an investigation method with confidence based on localized differential privacy. Background technique [0002] In real life, surveys are a widely used mechanism for collecting user data information. With the rapid development of information technology, online survey technology has gradually become the mainstream survey mode, which can facilitate more participants to participate in the survey. Surveys allow us to accurately collect useful information such as the distribution of participants' answers and opinions on specific topics, which helps us make decisions in applications such as market forecasting, election voting, and disease analysis. The traditional survey method assumes that the final survey answers submitted by the participants are exactly the same as the facts. In fact, the participants may not be sure about the answers to the survey questions they subm...

AI Technical Summary

Problems solved by technology

The literature [Borui Cai, Guangyan Huang, Chi-Hung Chi, Ran Liu and Yanfeng Shu. "AnAccurate Negative Survey Using Answer Confidence Level." IEEE Transactions on Dependable and Secure Computing, 2019, doi: 10.1109 / TDSC.2019.2941197] proposes the use of negative The survey method protects the participants' choices, but does not protect the confidence of the answers

Furthermore, to the best of our knowledge, there is currently no scheme that protects both participant answers and the corresponding confidence in survey methods with confidence

Images