CHARACTERIZING RESIDENTIAL BUILDING PATTERNS IN HIGHDENSITY CITIES USING GRAPH CONVOLUTIONAL NEURAL NETWORKS

Muxin Jia, Kaiheng Zhang, Taro Narahara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In urban morphology studies, accurately classify ingresidential building patterns is crucial for informed zoning and urban design guidelines. While machine learning, particularly neural networks, has been widely applied to urban form taxonomy, most studies focus on grid-like data from street-view images or satellite imagery. Our paper provides a novel framework for graph classification by extracting features of clustering buildings at different scales and training a spectral-based GCN model on graph-structured data. Furthermore, from the perspective of urban designers, we put forward corresponding design strategies for different building patterns through data visualization and scenario analysis. The findings indicate that GCNhas a good performance and generalization ability in identify ingresidential building patterns, and this framework can aid urban designers or planners in decision-making for diverse urban environments in Asia.

Original languageEnglish (US)
Title of host publicationAccelerated Design - 29th International Conference on Computer-Aided Architectural Design Research in Asia, CAADRIA 2024
EditorsNicole Gardner, Christiane M. Herr, Likai Wang, Hirano Toshiki, Sumbul Ahmad Khan
PublisherThe Association for Computer-Aided Architectural Design Research in Asia
Pages39-48
Number of pages10
ISBN (Print)9789887891826
DOIs
StatePublished - 2024
Event29th International Conference on Computer-Aided Architectural Design Research in Asia, CAADRIA 2024 - Singapore, Singapore
Duration: Apr 20 2024Apr 26 2024

Publication series

NameProceedings of the International Conference on Computer-Aided Architectural Design Research in Asia
Volume2
ISSN (Print)2710-4257
ISSN (Electronic)2710-4265

Conference

Conference29th International Conference on Computer-Aided Architectural Design Research in Asia, CAADRIA 2024
Country/TerritorySingapore
CitySingapore
Period4/20/244/26/24

All Science Journal Classification (ASJC) codes

  • Architecture
  • Building and Construction
  • Computer Graphics and Computer-Aided Design
  • Materials Science (miscellaneous)
  • Modeling and Simulation

Keywords

  • Building pattern classification
  • Graph convolutional neural networks
  • Machine learning
  • Urban morphology

Fingerprint

Dive into the research topics of 'CHARACTERIZING RESIDENTIAL BUILDING PATTERNS IN HIGHDENSITY CITIES USING GRAPH CONVOLUTIONAL NEURAL NETWORKS'. Together they form a unique fingerprint.

Cite this