Characterizing Key Players in Child Exploitation Networks on the Dark Net

  • Alain M. P. FonhofEmail author
  • Madeleine van der Bruggen
  • Frank W. Takes
Conference paper
Part of the Studies in Computational Intelligence book series (SCI, volume 813)


This paper studies online child exploitation networks in which users communicate about illegal child pornography material. Law enforcement agencies are extremely interested in better understanding these networks and their key players. We utilize unique real-world network data sets collected from two different online discussion forums on the dark net. Our study of the network structure underlying these forums results in three contributions.

First, we propose an approach to identify key players using various centrality measures, allowing us to automatically rank users. Experiments show that our method closely resembles a network-agnostic ranking of users created by domain experts. Second, network metrics are able to characterize a large portion of the users, allowing us to distinguish between regular users, managers and technical moderators. Finally, analyzing the structural properties and distributions of these networks in both the one-mode and two-mode perspective reveals various interesting network-driven insights, such as anti-lurker and anti-law enforcement policies and new user application guidelines. In addition, we found that active users form an elite that participate in more specialized discussions.


Criminal networks Key player identification Two-mode networks Child exploitation networks Network projection 


  1. 1.
    Barabási, A.L.: Network Science. Cambridge University Press, Cambridge (2016)zbMATHGoogle Scholar
  2. 2.
    Bonacich, P.: Power and centrality: a family of measures. Am. J. Sociol. 92(5), 1170–1182 (1987)CrossRefGoogle Scholar
  3. 3.
    Borgatti, S.P.: Identifying sets of key players in a social network. Comput. Math. Organ. Theory 12(1), 21–34 (2006)CrossRefGoogle Scholar
  4. 4.
    Brin, S., Page, L.: The anatomy of a large-scale hypertextual web search engine. Comput. Netw. ISDN Syst. 30(1–7), 107–117 (1998)CrossRefGoogle Scholar
  5. 5.
    Van der Bruggen, M., Blokland, A.: Child pornography and the internet: a systematic literature review. Under review (2018)Google Scholar
  6. 6.
    Duijn, P.: Detecting and disrupting criminal networks: a data driven approach. Ph.D. thesis, University of Amsterdam (2016)Google Scholar
  7. 7.
    Egan, M.: Thinking of venturing on to the dark web? You might want to change your mind. Tech Advisor (2018).
  8. 8.
    Latapy, M., Magnien, C., Del Vecchio, N.: Basic notions for the analysis of large two-mode networks. Soc. Netw. 30(1), 31–48 (2008)CrossRefGoogle Scholar
  9. 9.
    Newman, M.E.: Scientific collaboration networks. II. Shortest paths, weighted networks, and centrality. Phys. Rev. E 64(1), 016,132 (2001)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Nolker, R.D., Zhou, L.: Social computing and weighting to identify member roles in online communities. In: Proceedings of the 2005 IEEE/WIC/ACM International Conference on Web Intelligence, pp. 87–93. IEEE Computer Society (2005)Google Scholar
  11. 11.
    Westlake, B.G., Bouchard, M., Frank, R.: Finding the key players in online child exploitation networks. Policy Internet 3(2), 1–32 (2011)CrossRefGoogle Scholar
  12. 12.
    Ziegel, E.R.: Standard probability and statistics tables and formulae. Technometrics 43(2), 249 (2001)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Alain M. P. Fonhof
    • 1
    Email author
  • Madeleine van der Bruggen
    • 2
  • Frank W. Takes
    • 1
    • 3
  1. 1.Department of Computer Science (LIACS)Leiden UniversityLeidenThe Netherlands
  2. 2.Leiden Institute of Criminal Law and Criminology, Dutch National PoliceLeidenThe Netherlands
  3. 3.University of AmsterdamAmsterdamThe Netherlands

Personalised recommendations