Social network data differential privacy protection method based on uncertain graph

Abstract

The present invention discloses a social network data differential privacy protection method based on an uncertain graph. The method comprises: constructing an original graph, and constructing an uncertain graph with the weight value and an adjacency matrix of the uncertain graph according to the original graph; according to the uncertain graph, constructing a noise adjacency matrix which is required to be added by satisfying the differential privacy; and adding the adjacency matrix with the noise adjacency matrix of the uncertain graph to obtain the adjacency matrix of a to-be-published graph, converting the adjacency matrix of the to-be-published graph into a social network graph for publishing. According to the method disclosed by the present invention, when adding the noise, not only satisfying differential privacy is ensured, and more structure information of the original graph can be saved, so that the published graph will not add excessive noise, and the data analyst can do moreresearch and analysis in a facilitated manner.

Description

technical field [0001] The invention relates to the technical field of network data security, in particular to a social network data differential privacy protection method based on an uncertainty graph. Background technique [0002] With the development of the Internet, more and more social individuals register on social networks, and a large amount of personal information is collected and analyzed. This has also attracted the attention of many data researchers. The analysis of social networks can provide richer and more accurate data sources for data mining and pattern analysis. However, publishing social network data will lead to the leakage of sensitive information of users. In order to ensure the privacy and security of social individuals, privacy protection processing is required in the process of social network publishing. [0003] At present, the privacy protection results published on social networks can be mainly divided into three categories. One is the clustering...

AI Technical Summary

Problems solved by technology

[0004] What the present invention is to solve is the problem that the differential privacy protection method in the existing social network will add too much in the process of adding noise, and provides a social network data differential privacy protection method based on an uncertain graph. When adding noise, It not only guarantees the compliance with differential privacy, but also preserves more structural information of the original graph, so that the published graph will not add excessive noise, which is conducive to data analysts to do more research and analysis

Images