Social network analysis: the use of graph distances to compare artificial and criminal networks
De Meo, Pasquale
AffiliationUniversity of Palermo, Palermo, Italy.
University of Messina, Messina, Italy.
University of Derby
Free University of Bolzano, Bolzano, Italy.
MetadataShow full item record
AbstractAim: Italian criminal groups become more and more dangerous spreading their activities into new sectors. A criminal group is made up of networks of hundreds of family gangs which extended their influence across the world, raking in billions from drug trafficking, extortion and money laundering. We focus in particular on the analysis of the social structure of two Sicilian crime families and we used a Social Network Analysis approach to study the social phenomena. Starting from a real criminal network extracted from meetings emerging from the police physical surveillance during 2000s, we here aim to create artificial models that present similar properties. Methods: We use specific tools of social network analysis and graph theory such as network models (i.e., random, small-world and scale-free) and graph distances to quantify the similarity between an artificial network and a real one. To the best of our knowledge, spectral graph distances and the DeltaCon similarity have never been applied to criminal networks. Results: Our experiments identify the Barabási-Albert model as the one which better represents a criminal network. For this reason, we could expect that new members of a criminal organization will be more likely to establish connections with high degree nodes rather than low degree nodes. Conclusion: Artificial but realistic models can represent a useful tool for Law Enforcement Agencies to simulate and study the structure, evolution and faults of criminal networks.
CitationFicara, A., Curreri, F., Cavallaro, L., De Meo, P., Fiumara, G., Bagdasar, O. and Liotta, A., (2021). 'Social network analysis: the use of graph distances to compare artificial and criminal networks'. Journal of Smart Environments and Green Computing, 1(3), pp. 159-172.
PublisherOAE Publishing Inc.
JournalJournal of Smart Environments and Green Computing
Showing items related by title, author, creator and subject.
Space Wireless Sensor Networks for planetary exploration: Node and network architecturesRodrigues, Pedro; Oliveira, Andre; Oddi, Guido; Liberati, Francesco; Alvarez, Francisco; Cabas, Ramiro; Vladimirova, Tanya; Zhai, Xiaojun; Jing, Hongyuan; Crosnier, Michael; et al. (IEEE, 2014-07-14)Wireless Sensor Networks (WSNs) are made of a usually large number of nodes deployed over an area of interest in order to monitor specific phenomena. WSNs constitute a promising technology for planetary exploration, since they can be deployed in order to monitor the environmental conditions on a planet's surface, also in view of possible manned missions. This paper deals with the design of node and network architectures in a WSN targeted for planetary exploration, with a particular focus on the challenges, the driving principles and the design solutions adopted at WSN node and network level. Also, the paper introduces the basic architecture supporting data fusion at node and network level, which plays a fundamental role in order to increase overall WSN performances.
An exploratory social network analysis of academic research networksToral, Sergio L.; Bessis, Nik; Martinez-Torres, M. R.; Franc, Florian; Barrero, Federico; Xhafa, Fatos; University of Seville, Spain; University of Derby; University of Toulouse (INSA), France; Polytechnic University of Catalonia (UPC), Spain (IEEE, 2011-11)For several decades, academics around the world have been collaborating with the view to support the development of their research domain. Having said that, the majority of scientific and technological policies try to encourage the creation of strong inter-related research groups in order to improve the efficiency of research outcomes and subsequently research funding allocation. In this paper, we attempt to highlight and thus, to demonstrate how these collaborative networks are developing in practice. To achieve this, we have developed an automated tool for extracting data about joint article publications and analyzing them from the perspective of social network analysis. In this case study, we have limited data from works published in 2010 by England academic and research institutions. The outcomes of this work can help policy makers in realising the current status of research collaborative networks in England.
Artificial neural networks training acceleration through network science strategiesCavallaro, Lucia; Bagdasar, Ovidiu; De Meo, Pasquale; Fiumara, Giacomo; Liotta, Antonio; University of Derby; University of Messina, Polo Universitario Annunziata, 98122, Messina, Italy; Free University of Bozen-Bolzano, Bolzano, Italy (Springer Science and Business Media LLC, 2020-09-09)The development of deep learning has led to a dramatic increase in the number of applications of artificial intelligence. However, the training of deeper neural networks for stable and accurate models translates into artificial neural networks (ANNs) that become unmanageable as the number of features increases. This work extends our earlier study where we explored the acceleration effects obtained by enforcing, in turn, scale freeness, small worldness, and sparsity during the ANN training process. The efficiency of that approach was confirmed by recent studies (conducted independently) where a million-node ANN was trained on non-specialized laptops. Encouraged by those results, our study is now focused on some tunable parameters, to pursue a further acceleration effect. We show that, although optimal parameter tuning is unfeasible, due to the high non-linearity of ANN problems, we can actually come up with a set of useful guidelines that lead to speed-ups in practical cases. We find that significant reductions in execution time can generally be achieved by setting the revised fraction parameter (ζ) to relatively low values.