Clustering constrained on linear networks

TL;DR

This paper introduces a novel unsupervised clustering method for point events on linear networks, leveraging Dirichlet process-based random partition models to incorporate spatial effects and improve clustering accuracy.

Contribution

It develops a new clustering approach that accounts for spatial relationships on networks, with a Gibbs sampler for practical implementation and application to crime data.

Findings

Effective clustering of crime events on Mexico City network

Incorporation of spatial effects improves clustering accuracy

Sensitivity analysis validates the method's robustness

Abstract

An unsupervised classification method for point events occurring on a network of lines 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.