Network user structure disturbance method based on spectral constraint and sensitive area partition

Abstract

The invention belongs to the field of internet information safety, relates to a social network privacy protection method, and particularly relates to a network user structure disturbance method based on spectral constraint and sensitive area partition. The network user structure disturbance method includes the following steps: performing sensitive area and nonsensitive area partition for an input network user structure chart G, and then performing stochastic disturbance based on the spectral constraint on a social network graph, and outputting a network user structure chart G'. The network user structure disturbance method based on the spectral constraint and the sensitive area partition has the advantages that the usability of social network data is improved, invalid disturbance can be eliminated through the method, privacy protection degree is improved, and structural characteristics of a social network can be better maintained while privacy is protected.

Description

Technical field [0001] The invention belongs to the field of Internet information security, relates to a social network privacy protection method, in particular to a network user structure disturbance method based on spectrum constraints and sensitive area division. Background technique [0002] A network user is a point-like network topology composed of individuals or communities. Each point represents an individual, and there may be various interdependent social relationships between individuals, which are represented by edges between points in the topological network. This network topology is completely open on the Internet. Therefore, there is a risk of information leakage on the social network and the relationship between individuals, especially the sensitive information. In order to ensure the smooth analysis of social network data, it is required to maintain the structural characteristics of the original social network data. [0003] At present, the randomized pertu...

AI Technical Summary

Problems solved by technology

The privacy of social network users can be well protected as the number of additions, deletions, and exchange edges increases, but the availability of the network becomes poor after the disturbance

Another method is the randomized perturbation method based on spectral constraints. Compared with the randomized social network perturbation method, it can well maintain the structural characteristics of the social network, but the privacy protection is not strong enough.

Images