Research on Dynamic Social Network Anonymity Technology for Protecting Community Structure

Abstract

The dynamic change of vertex degree in a dynamic social network will lead to vertex identity disclosure given the deficiencies in current privacy protection methods, such as the destruction of community structure and low data processing capability of a single workstation. The dynamic social network degree sequence anonymity (DSNDSA) method to protect community structure is proposed. The method obtains the grouping and anonymous results based on a compressed binary tree constructed by a new method called a multidimensional vector. Dummy vertices are added in parallel to construct anonymous graphs. Distributed to merge dummy vertices method based on the community is designed to reduce the number of vertices added to satisfy the anonymity model. A divide and the agglomerate algorithm is expanded for community detection. The experimental results show that the proposed algorithm based on GraphX can overcome the defects of the traditional algorithm in community protection while meeting the requirement of anonymity.