Graph Comparison and Artificial Models for Simulating Real Criminal Networks

Cavallaro, Lucia; Ficara, Annamaria; Curreri, Francesco; Fiumara, Giacomo; Meo, Pasquale De; Bagdasar, Ovidiu; Liotta, Antonio

doi:10.1007/978-3-030-65351-4_23

Cited by 9 publications

(10 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In both networks, nodes are suspected criminals and edges represent meetings (MN), or recorded phone calls (PC). These original datasets have been already studied in some of our previous works [ 10 , 16 – 18 , 29 , 45 ] and they are available on Zenodo [ 35 ].…”

Section: Methodsmentioning

confidence: 99%

“…In the attempt of gaining a deeper understanding of criminal networks, in our previous work [ 29 ] we used graph distances to compare randomly generated graphs and a real criminal network. In the present paper, we have implemented distances, to understand the extent by which a partial knowledge of a criminal network may negatively affect the investigations.…”

Section: Methodsmentioning

confidence: 99%

“…Some reviews on the most common graph comparison metrics are [25][26][27][28]. In [29], such distance measures were exploited to quantify how well artificial (but realistic) models can simulate real criminal networks. The same measures are used herein for a different task.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Criminal networks analysis in missing data scenarios through graph distances

Ficara

Cavallaro²,

Curreri³

et al. 2021

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Data collected in criminal investigations may suffer from issues like: (i) incompleteness, due to the covert nature of criminal organizations; (ii) incorrectness, caused by either unintentional data collection errors or intentional deception by criminals; (iii) inconsistency, when the same information is collected into law enforcement databases multiple times, or in different formats. In this paper we analyze nine real criminal networks of different nature (i.e., Mafia networks, criminal street gangs and terrorist organizations) in order to quantify the impact of incomplete data, and to determine which network type is most affected by it. The networks are firstly pruned using two specific methods: (i) random edge removal, simulating the scenario in which the Law Enforcement Agencies fail to intercept some calls, or to spot sporadic meetings among suspects; (ii) node removal, modeling the situation in which some suspects cannot be intercepted or investigated. Finally we compute spectral distances (i.e., Adjacency, Laplacian and normalized Laplacian Spectral Distances) and matrix distances (i.e., Root Euclidean Distance) between the complete and pruned networks, which we compare using statistical analysis. Our investigation identifies two main features: first, the overall understanding of the criminal networks remains high even with incomplete data on criminal interactions (i.e., when 10% of edges are removed); second, removing even a small fraction of suspects not investigated (i.e., 2% of nodes are removed) may lead to significant misinterpretation of the overall network.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Criminal networks analysis in missing data scenarios through graph distances

Ficara

Cavallaro²,

Curreri³

et al. 2021

PLoS ONE

Self Cite

View full text Add to dashboard Cite

show abstract

“…From the analysis of this document, we initially built two graphs: Meetings and Phone Calls, in which nodes were uniquely associated with suspected criminals and edges specified meetings and phone calls respectively among individuals [14][15][16][17]. The Meetings graph had 101 nodes and 256 edges.…”

Section: Datasetmentioning

confidence: 99%

“…We propose a new real multilayer criminal network derived from two simple graphs described in our earlier works [14][15][16][17], whose datasets are publicly available on Zenodo [18]. These simple graphs capture the physical meetings and phone calls among suspects in an anti-mafia operation called "Montagna", concluded in 2007 by the Public Prosecutor's Office of Messina (Sicily).…”

Section: Introductionmentioning

confidence: 99%

Multilayer Network Analysis: The Identification of Key Actors in a Sicilian Mafia Operation

Ficara,

Fiumara,

De Meo

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, Social Network Analysis studies have led to an improvement and to a generalization of existing tools to networks with multiple subsystems and layers of connectivity. These kind of networks are usually called multilayer networks. Multilayer networks in which each layer shares at least one node with some other layer in the network are called multiplex networks. Being a multiplex network does not require all nodes to exist on every layer. In this paper, we built a criminal multiplex network which concerns an anti-mafia operation called "Montagna" and it is based on the examination of a pre-trial detention order issued on March 14, 2007 by the judge for preliminary investigations of the Court of Messina (Sicily). "Montagna" focus on two Mafia families called "Mistretta" and "Batanesi" who infiltrated several economic activities including the public works in the north-eastern part of Sicily, through a cartel of entrepreneurs close to the Sicilian Mafia. Originally we derived two single-layer networks, the former capturing meetings between suspected individuals and the latter recording phone calls. But some networked systems can be better modeled by multilayer structures where the individual nodes develop relationships in multiple layers. For this reason we built a two-layer network from the single-layer ones. These two layers share 47 nodes. We followed three different approaches to measure the importance of nodes in multilayer networks using degree as descriptor. Our analysis can aid in the identification of key players in criminal networks.

show abstract