Research Article
Differentially Private Social Graph Publishing for Community Detection
2 downloads
@INPROCEEDINGS{10.1007/978-3-030-63095-9_11, author={Xuebin Ma and Jingyu Yang and Shengyi Guan}, title={Differentially Private Social Graph Publishing for Community Detection}, proceedings={Security and Privacy in Communication Networks. 16th EAI International Conference, SecureComm 2020, Washington, DC, USA, October 21-23, 2020, Proceedings, Part II}, proceedings_a={SECURECOMM PART 2}, year={2020}, month={12}, keywords={Differential privacy Community detection Social network}, doi={10.1007/978-3-030-63095-9_11} }- Xuebin Ma
Jingyu Yang
Shengyi Guan
Year: 2020
Differentially Private Social Graph Publishing for Community Detection
SECURECOMM PART 2
Springer
DOI: 10.1007/978-3-030-63095-9_11
Abstract
Social networks typically include a community structure, and the connections between nodes within the same community are very close; however, the connections between communities are sparse. In this study, we analyze the main challenges behind the problem and then resolve it using differential privacy. First, we choose the Louvain algorithm as a benchmark community detection algorithm for the algorithmic perturbation scheme. We introduce an exponential mechanism that uses modularity as a score. Secondly, by transforming each community into a hierarchical random graph model, and its edge connection probability is noisy by differential privacy mechanism to ensure the security of relevant information in the protected community.
Copyright © 2020–2025 ICST
