Privacy-Preserving Graph Operations for Social Network Analysis

Abstract

Nowadays, our daily life is surrounded by various social networks, and they play an important role for people to communicate with others. The social networks contain large amount of valuable information, that can be used for research and business purposes. As a result, social network analysis and data mining receive lots of research attentions in recent years. Graph structure is commonly used in social network analysis, since it is easy to convert the data in social networks into graph-structured data, and various graph algorithms can help to solve different computing problems. In this paper, we investigate performing graph operations in a privacy-preserving manner, which are widely used in social network analysis. We propose two protocols that allow two parties to jointly compute the intersection and union of their graphs. Our protocols utilize homomorphic encryption to prevent information leakage during the process, and we provide security proofs of the protocols in the semi-honest setting.