GeoClustering.jl

Geostatistical clustering methods for the GeoStats.jl framework.

SLIC

Simple Linear Iterative Clustering (SLIC) produces clusters that are spatially connected based on a geospatial distance dₛ. The samples in these clusters are similar in terms of their features according to a distance dᵥ. The tradeoff is controlled with a hyperparameter parameter m in an additive model dₜ = √(dᵥ² + m²(dₛ/s)²). The original method developed for images is described in Achanta et al. 2011. It has been generalized in this package for any geospatial data set (e.g. point sets).

GHC

Geostatistical Hierarchical Clustering (GHC) produces clusters based on (cross-)variograms between covariates and on a kernel function between geospatial coordinates. The method is described in Fouedjio, F. 2016.

GSC

Geostatistical Spectral Clustering (GSC) produces clusters based on the spectral decomposition of the graph Laplacian constructed with weights that are a function of features and locations. The method is described in Romary et al. 2015.

Installation

Get the latest stable release with Julia's package manager:

] add GeoClustering

Usage

This package is part of the GeoStats.jl framework.

For a simple example of usage, please check the main documentation.

Asking for help

If you have any questions, please contact our community.