Auto-Contractive Maps and Minimal Spanning Tree: Organization of Complex Datasets on Criminal Behavior to Aid in the Deduction of Network Connectivity
Using the Metropolitan Police Service Central Drug Trafficking Database (London, UK), the Auto-Contractive Map neural network technology is applied to identify unknown associations among individuals contained within it. Whilst a database may contain associations of known individuals, it may be mined to identify associations that are altogether unknown and unexpected. It is shown that individuals can be identified who belong to the same gang or drug trafficking circle. The results produce a profile that is mathematically justified and devoid of any political involvement. Key to this analysis is the organization of the datasets such that a meaningful analysis and interpretation can be made. This organization is described and illustrated using the drug trafficking data.
KeywordsDrug Trafficking Police Operation Previous Offense Drug Seizure Previous Arrest
- Massimo, B., & Sacco, P. L. (2010). Auto-contractive maps, the H function, and the maximally regular graph (MRG): a new methodology for data mining (chapter 11). In V. Capecchi et al. (Eds.), Applications of mathematics in models, artificial neural networks and arts. Dordrecht/London: Springer. doi: 10.1007/978-90-481-8581-8_11.Google Scholar
- Buscema, M. (2007). Contractive Maps. Software for programming Auto Contractive Maps (Semeion Software #15, v. 2), Rome.Google Scholar
- Buscema, M. (2007). Constraints Satisfaction Networks. Software for programming Non Linear Auto-Associative Networks, Semeion Software #14 (v. 10), Rome.Google Scholar
- Buscema, M. (2008). MST. Software for programming Trees from artificial networks weights matrix (Semeion Software #38, v 5), Rome.Google Scholar
- Massini, G. (2007). Tree Visualizer. Software to draw and manipulate tree graph (Semeion Software #40, v. 3), Rome.Google Scholar