Longitudinal Analysis of the Topology of Criminal Networks using a Simple Cost-Benefit Agent-Based Model
conference paper
Recently, efforts have been made in computational criminology to study the dynamics of criminal organisations and improve law enforcement measures. To understand the evolution of a criminal network, current literature uses social network analysis and agent-based modelling as research tools. However, these studies only explain the short-term adaptation of a criminal network with a simplified mechanism for introducing new actors. Moreover, most studies do not consider the spatial factor, i.e. the underlying social network of a criminal network and the social environment in which it is active. This paper presents a computational modelling approach to address this literature gap by combining an agent-based model with an explicit social network to simulate the long-term evolution of a criminal organisation. To analyse the dynamics of a criminal organisation in a population, different social networks were modelled. A comparison of the evolution between the different networks was carried out, including a topological analysis (secrecy, flow of information and size of largest component). This paper demonstrates that the underlying structure of the network does make a difference in its development. In particular, with a preferentially structured population, the prevalence of criminal behaviour is very pronounced. Moreover, the preferential structure provides criminal organisations a certain efficiency in terms of secrecy and flow of information
TNO Identifier
987548
Publisher
Springer Nature
Source title
International Conference on Computational Science
Place of publication
Cham
Pages
10-24
Files
To receive the publication files, please send an e-mail request to TNO Repository.