Privacy-Preserving Group Closeness Maximization

This study explores the metric of group closeness centrality within the framework of social networks, a departure from the traditional analysis focused solely on the significance of individual nodes. Given the intricate dynamics observed in networks governed by various stakeholders, we introduce a framework that preserves privacy through the application of a greedy algorithm. This approach is designed to evaluate the collective influence of groups while ensuring the confidentiality of individual data. Furthermore, we employ Oblivious Random Access Memory (ORAM) [1] within cloud servers to conceal access patterns, thereby enhancing data privacy. Through comprehensive experimentation across three real-world social network datasets within the MP-SPDZ framework [2], dedicated to secure multi-party computation, we demonstrate the efficiency of our proposed methods.

I. Introduction

Social networks constitute social structures comprising various social actors and their interactions. Conceptually, these can be modeled as graph data structures, with nodes representing actors and edges symbolizing interactions between them [3], [4]. Social Network Analysis (SNA) has rapidly evolved into a forefront discipline, attracting significant research interest. Within SNA, centrality quantifies the extent to which a node, or actor, occupies a central position in the network, with centrality scores numerically representing this measure. A multitude of methods for measuring centrality have been developed, addressing various research goals. Yet, most studies concentrate on assessing centrality at the individual node level [5]-[7]. In contrast, practical applications often necessitate identifying a group of k nodes that collectively exhibit the highest closeness centrality, beyond simply generating a top-k ranking. This approach to centrality measurement, by focusing on node sets, offers a more precise depiction of the most centrally positioned groups within a social network.