Social network data privacy protection method based on graph primitives

Abstract

The invention discloses a social network data privacy protection method based on graph primitives. After data are initialized, an original unweighted graph is transformed into a weighted graph, and degrees of graph nodes are arranged in a descending and subjected to a K-degree anonymity algorithm to obtain a descending anonymity degree sequence. The difference between the anonymity degree sequenceand an original degree sequence is acquired, nodes needing edge modification are classified according to difference values and combined into a candidate set, the candidate set is modified according to corresponding selection standards until anonymity requirements are met, and anonymous social network data are released. When the network data are released, a graph primitive structure in a network is effectively reserved while the anonymity requirements are met, and related data analysis of data analyzers is facilitated.

Description

technical field [0001] The invention relates to the technical field of data privacy protection, in particular to a method for protecting social network data privacy based on graphlets. Background technique [0002] In recent years, social networking has become more and more popular, so that more and more people start using online social networking to communicate with friends, family, and colleagues. It is precisely because many people share some personal information through social networks that social networks have become an important source of data for research and mining in many fields. Many users' private information is embedded in the data, so the data owner should avoid leaking some private and sensitive information of the users when releasing the data. Therefore, anonymity processing becomes particularly important. [0003] Currently, social network analysis is only achieved by capturing low-level structures (nodes and edges) in network graphs. However, some networks...

AI Technical Summary

Problems solved by technology

[0004] The traditional simple anonymous privacy protection technology is no longer enough to protect the privacy of users. In order to better protect the privacy of data in social networks, the more popular anonymous technologies are: node K-anonymity, that is, each node is at least K-1 Other nodes are indistinguishable, so the probability of each node being successfully identified does not exceed 1 / K; K-degree anonymity, that is, assuming that the attacker knows the degree (edge) information of all nodes, at least K-1 nodes are indistinguishable after anonymization

Images