Privacy protection method and system oriented to large-scale graph data dissemination

Abstract

The invention discloses a privacy protection method oriented to large-scale graph data dissemination. The privacy protection method specifically comprises the steps of 1, uniformly dividing original graph data into multiple subsidiary blocks; 2, reading cut connection sides, comparing node degrees of the two ends of each cut connection side, newly adding noise nodes in a subsidiary block of each node with a large node degree, and reserving the cut connection sides through a ligature of the corresponding node with the large node degree and the corresponding noise node; 3, constructing an isomorphism block matrix, wherein subsidiary block structure information is compared with the isomorphism block matrix, and isomorphism is conducted in the mode of adding the noise sides to complete graph data anonymous protection. The time needed for the overall graph data anonymous protection is lowered by one order of magnitudes, and the efficiency of an anonymization process is achieved; finally ananonymous graph satisfies a k anonymous mechanism, and the safety of anonymization is achieved. By means of the privacy protection method oriented to the large-scale graph data dissemination, it is guaranteed that the balance performance of the anonymous graph among usability, safety and efficiency is improved by a large margin.

Description

technical field [0001] The invention belongs to the technical field of graph data processing, and more specifically relates to a privacy protection method and system for publishing large-scale graph data. Background technique [0002] In recent years, with the explosive growth of data shared by multiple parties, both academia and industry are in urgent need of large-scale data processing technologies, especially for unstructured data such as large-scale graph data. Due to the structural characteristics of graph data, it is widely used in many data storage and analysis fields. For example, Facebook uses the TAO system to store its large-scale social network data. However, such social network data contains a large amount of user personal privacy information, and the data owner needs to perform special privacy protection processing on the original data before releasing and sharing the data to third parties. Generally speaking, simply remove the user ID from the original image...

AI Technical Summary

Problems solved by technology

However, the existing privacy protection scheme based on the k-anonymity mechanism has the following outstanding problems: 1. Low efficiency and long anonymization cycle: because the existing technology spends a lot of time in the process of graph segmentation, the overall efficiency of the anonymity scheme is greatly reduced ;2. Poor usability: The usability of existing technologies is limited by the advantages and disadvantages of graph segmentation. Due to the uncertainty of graph segmentation, the anonymous process often needs to add a lot of noise information and delete information on the original image, making the availability of anonymous graphs random. 3. Poor balance: Existing anonymous schemes often focus on the availability or efficiency of the scheme, and lack the guarantee for the dynamic balance among usability, efficiency and security

Images