# GeoGraphNetworks: A comprehensive benchmark dataset for accurate and scalable graphical representations

## TL;DR

GeoGraphNetworks provides ready-to-use spatial networks for the US and UK, simplifying research in transportation, urban planning, and network science.

## Contribution

The paper introduces a validated, publicly accessible benchmark dataset of spatial networks in usable graph formats.

## Key findings

- The dataset includes 110 validated spatial networks covering US road and rail, and UK road and river networks.
- Each network contains node coordinates, edge connectivity, and edge lengths in kilometers.
- GeoGraphNetworks reduces preprocessing needs and supports multilingual programming environments.

## Abstract

Accurate graphical representations of real-world systems are essential for research in fields such as transportation, urban planning, ecology, and network science. While ESRI Shapefiles are a widely used source of geospatial vector data, converting them into topologically correct and usable network format requires significant technical expertise and computational resources. To address this challenge, we present the GeoGraphNetworks data repository, which contains 110 validated spatial networks and offers comprehensive and publicly accessible spatial network resources for the road and rail networks of the United States of America and the road and river networks of Great Britain. This data repository eliminates the need for users to perform complex geospatial processing by providing workable graph representations in JSON and XLSX formats. Each network includes detailed geographic information such as node coordinates (latitude and longitude), edge connectivity, and edge lengths in kilometres. By reducing preprocessing overhead and enabling immediate application across multilingual programming environments, GeoGraphNetworks lowers technical barriers and supports reproducible, scalable spatial network research across disciplines.