Identifying criminal organizations from their social network structures

Abstract

Identification of criminal structures within very large social networks is an essential security feat. Byidentifying such structures, it may be possible to track, neutralize, and terminate the corresponding criminal organizationsbefore they act. We evaluate the effectiveness of three different methods for classifying an unknown network as terrorist,cocaine, or noncriminal. We consider three methods for the identification of network types: evaluating common socialnetwork analysis metrics, modeling with a decision tree, and network motif frequency analysis. The empirical resultsshow that these three methods can provide significant improvements in distinguishing all three network types. We showthat these methods are viable enough to be used as supporting evidence by security forces in their fight against criminalorganizations operating on social networks.