PostGeoDa (beta)

Xun Li (@lixun910), Luc Anselin

Run spatial data analysis simply using SQL in your PostgreSQL database!! 🚀

Introduction

PostGeoDa is a PostgreSQL extension for spatial data analysis. PostGeoDa is developed using C/C++ based on the libgeoda library. By utilizing and the database architecture of PostgreSQL and the spatial index in PostGIS, PostGeoDa has the ability to handle real big spatial data.

-- Create Queen contiguity weights
SELECT queen_weights(gid, the_geom) OVER() FROM natregimes;

-- Apply local Moran statistics
SELECT local_moran(hr60, queen_weights) OVER() FROM natregimes;

-- Apply spatial regionalization SKATER
SELECT skater(ARRAY[hr60, dv60, ue60], queen_weights) OVER() FROM natregimes;

PostGeoDa is a free and open-sourced library. It is released under the GNU General Public License (GPLv2 or later). PostGeoDa is developed by Xun Li and Luc Anselin.

Why PostGeoDa? 🤔

• PostGeoDa is the first spatial data analysis extension of PostgreSQL to power spatial data services for cloud mapping platforms.

• PostGeoDa offers many features of spatial data analysis that run efficiently in PostgreSQL, so there is no need to spend extra time transferring geometries over the network.

• PostGeoDa has no dependencies. But it is designed to work with PostGIS to handle big spatial data.

• PostGeoDa works seamlessly with the current SQL API frameworks e.g. CARTO SQL API

If you prefer running spatial data analysis in browser, please check out jsgeoda. If you prefer running spatial data analysis in Python, please check out pygeoda. If you prefer running spatial data analysis in R, please check out rgeoda.

Contents

Version 0.0.1

• 0 Installation

• 1 Choropleth Mapping

  • Basic Mapping

  • Rate Mapping

  • Spatial Rate Mapping

• 2 Spatial Weights

  • Contiguity-Based Weights

  • Distance-Based Weights

  • Kernel Weights

• 3 Local Spatial Autocorrelation

  • Local Moran

  • Local Geary

  • Local Getis-Ord G

  • Local Join Count

  • Quantile LISA

• 4 Local Spatial Autocorrelation - Multivariate

  • Local Neighbor Match Test

  • Multivariate Local Geary

  • Bivariate Local Join Count

  • Multivariate Local Join Count

  • Multivariate Quantile LISA

• 5 Spatial Clustering

  • SKATER

  • REDCAP

• 6 Cluster Analysis

  • HDBSCAN

  • Fast K-Medoids