Clustering constrained on linear networks

An unsupervised classification method for point events occurring on a geometric network is proposed. The idea relies the distributional flexibility and practicality of random partition models to discover clustering structure featuring observations from particular phenomenon taking place given set edges. By incorporating spatial effect in distribution, induced by Dirichlet process, one able control distance between edges events, thus leading an appealing method. A Gibbs sampler algorithm proposed evaluated with sensitivity analysis. proposal motivated illustrated analysis crime violence patterns Mexico City.