Social network graph acquisition method and system and electronic equipment

Abstract

The invention relates to the technical field of security, and provides a social network graph acquisition method and system and electronic equipment, which are used for performing community division on an undirected graph corresponding to a plurality of preset users to obtain a plurality of sub-graphs, and adding Gaussian noise to each sub-graph for disturbance, so that the problem of privacy exposure in a social network graph is solved. Therefore, for an attacker, even if all information of the social network graph and related background knowledge of the original social network are obtained, the social network graph cannot be attacked; according to the technical scheme, the privacy and security of the social network graph are improved, the structural attributes of the social network graph are kept, and the availability of the social network graph is improved.

Description

technical field [0001] The present invention relates to the field of security technology, in particular to a method, system and electronic equipment for acquiring a social network graph. Background technique [0002] Nowadays, social network data is growing exponentially, and social network data is used in existing technologies such as information science, information dissemination, marketing, and emotional flow. However, from the user's point of view, social network data contains sensitive information of the user, which can be understood as private information. When a social network graph is generated and published based on social network data, an attacker can attack the social network graph. These sensitive information, that is, private information, can be mined out to cause adverse effects. Therefore, before the social network graph is released, how to process it so that the user's privacy can be effectively protected has become an important topic in this research field....

AI Technical Summary

Problems solved by technology

[0003] At present, there are a variety of technologies used to protect privacy in social networks, such as k-anonymity technology, perturbation algorithm, link perturbation, differential privacy, etc., but there are defects. For example, in 1998, Samarati and Sweeney proposed k-anonymity technology, The main idea is that, for the quasi-identifier attribute value of each tuple in the published data table, there are at least k-1 tuples with the same quasi-identifier attribute value, k is a positive integer greater than 1, such that The attacker cannot identify the target object through multi-table link analysis

Although this method can protect the links in the network, it is easy to destroy the structural properties of the social network graph, which is difficult to use in practical applications, and the privacy of the social network graph after disturbance is low

Images