This paper introduces Urbanity, a Python package for automated construction of feature-rich urban networks. It leverages open data to supplement urban networks with location information, enabling more expressive urban machine-learning models. The paper discusses data sources, software features, and presents network data from five global cities. Experiments show that adding contextual features significantly improves the accuracy of classifying different connection types within a network.
Publisher
npj Urban Sustainability
Published On
Jul 25, 2023
Authors
Winston Yap, Rudi Stouffs, Filip Biljecki
Tags
Urbanity
Python package
urban networks
open data
machine learning
contextual features
network data
Related Publications
Explore these studies to deepen your understanding of the subject.