LAUSR

In this paper, we introduce a new approach to the analysis and prediction of the spatial pattern of terrorist attacks. Our analysis makes use of a new methodology that is based upon a topological rather than geometric approach. It exploits the space grammar of the observed events to derive complex inferences which relate to the deep structural layers of the data generating process, and to the organizational architecture of its spatial distribution. We test our methodology on two of the currently most important and virulent theaters of terrorist activity, Libya and Syria, and we find that our methodology provides a rich array of insights which cannot be expected to be generated by more traditional geographical profiling methods.