Clustering constrained on linear networks

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

Automated summary

An unsupervised classification method is proposed for point events occurring on a geometric network. It utilizes the flexibility and practicality of random partition models to discover clustering structures of observations from a specific phenomenon on a given set of edges. By incorporating spatial effects through a random partition distribution induced by a Dirichlet process, the method offers an appealing clustering approach. A Gibbs sampler algorithm is proposed and evaluated with sensitivity analysis. The analysis of crime and violence patterns in Mexico City serves as the motivation and illustration for this proposal.