Social networks integration and privacy preservation using subgraph generalization

Abstract

Intelligence and law enforcement force make use of terrorist and criminal social networks to support their investigations such as identifying suspects, terrorist or criminal subgroups, and their communication patterns. Social networks are valuable resources but it is not easy to obtain information to create a complete terrorist or criminal social network. Missing information in a terrorist or criminal social network always diminish the effectiveness of investigation. An individual agency only has a partial terrorist or criminal social network due to their limited information sources. Sharing and integration of social networks between different agencies increase the effectiveness of social network analysis. Unfortunately, information sharing is usually forbidden due to the concern of privacy preservation. In this paper, we introduce the KNN algorithm for subgraph generation and a mechanism to integrate the generalized information to conduct social network analysis. Generalized information such as lengths of the shortest paths, number of nodes on the boundary, and the total number of nodes is constructed for each generalized subgraphs. By utilizing the generalized information shared from other sources, an estimation of distance between nodes is developed to compute closeness centrality. Two experiments have been conducted with random graphs and the Global Salafi Jihad terrorist social network. The result shows that the proposed technique improves the accuracy of closeness centrality measures substantially while protecting the sensitive data.