- 期刊
Personalized K-In&Out-Degree Anonymity Method for Large-scale Social Networks Based on Hierarchical Community Structure
摘要
The existing social network privacy protection technologies have the problems of neglecting the protection of the community structure and failing to meet users' different privacy protection requirements when processing the large-scale social network directed graphs. A personalized K-In&Out-Degree anonymity (PKIODA) algorithm based on hierarchical community structure is proposed. The algorithm divides the community based on the hierarchical community structure. According to the different user privacy protection levels Lv0~Lv3, the grouping and the anonymity sequence are distributed, and the nodes are added in parallel to realize the anonymity. Based on GraphX to transfer information between nodes, the virtual node pairs are merged and deleted according to the hierarchical community entropy change to reduce the information loss. The experimental results show that the PKIODA algorithm improves the efficiency of processing large-scale social network directed graph data, ensures the high availability of community structure analysis during data release, and satisfies the privacy protection requirements of different users.