Visualising the visibility graph analysis: A comparative analysis of VGA metrics representation and significance at micro-urban scale

Mara, Federico, Altafini, Diego ORCID: https://orcid.org/0000-0002-6559-2372, Salardi-Jost, Monique and Cutini, Valerio 2024. Visualising the visibility graph analysis: A comparative analysis of VGA metrics representation and significance at micro-urban scale. Presented at: 14th Space Syntax Symposium, Nicosia, Cyprus, 24-28 June, 2024. Item availability restricted.

PDF - Accepted Post-Print Version Restricted to Registered users only until 31 August 2024 due to copyright restrictions. Download (1MB) | Request a copy

Abstract

Visibility Graph Analysis (VGA) demonstrated its potential in extracting intrinsic spatial characteristics and in interpreting social behaviour in micro-urban scale scenarios. Given its granularity, it tends to perform better in capturing pedestrian movement and visual-based patterns in irregular spaces, which are not well apprehended through Isovist-decomposition-based methods. Despite these potentialities, VGA still presents issues, associated with its dependence on a superimposed grid, which often leads to depth-based distortions; which require mathematical investigations towards value normalization. Still, another aspect that requires investigation, concerns visualization methods, as better fitting them to the data distribution can also improve the effectiveness of VGA Analysis. Considering this, this approach proposes a methodological discussion, confronting the default visualization method used in the Space Syntax framework (Equal Interval) and a data distribution method (Equal Count) that is better suited to represent ordinal-level data – often associated with network measures. The objective is to highlight how Space Syntax’ VGA can visualise different micro-urban dynamics and how these visualisations inform the underlying system hierarchies, that are often unseen through the DepthMapX default data overview. VGA measures are statistically tested for normality, then histograms are constructed to visualize their data distributions, and set alongside respective spatializations. Discussions highlight patterns that emerge from each data distribution method. The study can then serve as a guide in selecting which representations for these metrics to use and how to interpret such visual graphs, thus providing a tool for enhancing the effectiveness of VGA and understanding logics within urban environment at a micro-urban scale.

Item Type:	Conference or Workshop Item (Paper)
Status:	In Press
Schools:	Architecture
Funders:	This project has received funding from the United Kingdom Research and Innovation Post Doctoral Fellowship Guarantee Scheme, set over the European Union’s Horizon Europe – Marie Skłodowska Curie Actions Post Doctoral Fellowships. UKRI Grant no. 101107846-
Date of First Compliant Deposit:	8 May 2024
Last Modified:	19 Jul 2024 13:22
URI:	https://orca.cardiff.ac.uk/id/eprint/168776

Actions (repository staff only)

Edit Item