A privacy-preserving intersection calculation method

Abstract

The invention provides an intersection calculation method for privacy protection, which comprises the following steps: (1) after performing privacy protection treatment on a set data by a client A and a client B, submitting to a server S, wherein the privacy protection treatment comprises the steps of calculating a discrete power of each element in the set, multiplying by a specific random digit, and submitting the number and product to the server S; (2) calculating an intersection calculation result by the server S, and respectively feeding the result back to the client A and the client B, specifically, dividing the elements in the acquired anonymous set, comparing with a specific threshold value, reserving the elements in the threshold value, and respectively sending the number to the client A and client B by the server S. According to the intersection calculation method for privacy protection provided by the invention, the defects of the prior art are overcome, the content privacy and the linkage privacy can be protected, the fuzzy intersection can be calculated, the calculated amount is small and the calculation is easily realized.

Description

technical field [0001] The invention relates to a basic calculation method in the field of mobile Internet security, in particular to a data intersection calculation method capable of protecting privacy. This method can be applied to privacy protection in e-commerce, social network, big data and other fields. Background technique [0002] In the mobile Internet, there are a large number of cases where the intersection of the sets of two clients is compared, such as the topics that two clients are concerned about in Weibo, the friends of two clients in social networks, and the interests of two clients in e-commerce. However, when calculating the intersection of two sets, the privacy of the client needs to be protected. Therefore, a perfect solution to this problem needs to meet four conditions: [0003] (1) The elements in the two collections submitted by the client should be anonymized, and the content of the collection should be encrypted or protected. This kind of privacy...

AI Technical Summary

Problems solved by technology

[0004] (2) The elements in the collection submitted to the server should be one-off and cannot be reproduced. If repeated, it will lead to a kind of privacy violation, called link privacy violation, such as repeated purchases, friend reproduction, etc.;

[0006] (4) The amount of calculation should not be large, and the efficiency should be high, because this calculation is very frequent

[0007] However, the existing methods to solve this problem are either inefficient and computationally complex; or cannot calculate fuzzy intersection; or methods cannot protect link privacy; or cannot protect content privacy