Analyzing and Responding to Resilience in Criminal Networks
The paper entitled “Resilience in Criminal Networks” applies network and graph theory to research on the resilience of criminal networks. This paper makes use of social network analysis (SNA) to explore how criminal networks respond to disruptive events and to make predictions about future behaviors in the network given external or internal factors. This research is aimed at bettering law enforcement strategy by suggesting specific nodes in a network whose removal would most disrupt the criminal organization. This paper relates to topics we have discussed in class on network analysis, as topics like density, clustering coefficients, local bridges, and components are used in analysis of the criminal network and to recommend better strategic law enforcement responses to criminal organizations.
According to this paper, the resilience of a criminal network describes the ability of the criminal network to face pressures from law enforcement agencies and rapidly reorganize after destabilizing attacks. The paper finds that criminal networks with high levels of resilience structure their network in two ways: First, they manage the flow of information by dividing members’ competencies into different components or subgroups, so as to prevent exposure of the entire network, and second, they allow for redundant nodes, so that the removal of a single node (criminal being arrested) minimizes impact to the exchange of information and resources in the network.
Visualization and subsequent analysis of a criminal network in graph form can likewise reveal ways to disrupt the network. By imagining criminal networks as a set of nodes (people, buildings, weapons, and other resources) connected by a set of edges (communication or links between nodes), law enforcement can then reevaluate strategies to combat the criminal network. By leveraging analysis of local bridges and highly clustered nodes, law enforcement can make decisions about which individuals or resources in the network to tackle first, while also being able to predict the responses/behavior of the network to these actions. Analysis can likewise reveal the density of a network (the level of cohesion in the network or the number of actual connections compared to the total possible connections), the centrality of a network (how concentrated a network is), and the clustering coefficient of specific nodes, which provides additional insights on how to leverage limited law enforcement resources.
In short, the topics we learned in class have real life applications to understanding and responding to criminal networks. Concepts like density, centrality, local bridges, and clustering coefficients can be used to make decisions about how to response to a criminal network and predict its behaviors.
Link: http://cab.unime.it/journals/index.php/AAPP/article/view/AAPP.942A1/AAPP942A1