AI-Integrated Pattern Prediction Model: Exploring the Influence of Building Fabric Morphology on Land Use Pattern

Date

2025-07-16

Author

Karadoğan, Selen
Efeoğlu, Hulusi Eren
Yetkin, Ozan
Çoban, Aybüke Balahun

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

52
views

0
downloads

Cite This

Morphological analysis using AI is anemerging but rapidly evolving field in urbanstudies, whereas analyzing urban tissues todifferentiate the characteristics of variouscities has long been a core approach inurban morphology. The relationship betweenform and function remains central to urbanmorphology discussions; however, thequantitative and dynamic application of AIacross different urban patterns remainsunexplored. In this context, this study exploresthe influence of morphological metrics of thebuilding fabric on land use patterns acrossdifferent urban contexts. It aims to investigatethe relationship between land use and its mostrelevant morphological elements by analyzing3D spatial data using an AI- driven model.The proposed method extracts 3D datafrom OpenStreetMap and processes it withclassification methods using Scikit Learn. Itleverages land use types using morphologicalfeatures by using the building as a unit ofanalysis including frontage ratio, height/width ratio, built-up coverage, compactness,accessible number of neighboring buildings,closeness, and betweenness centralities topredict land use types. The model will bedeveloped by using 10 km radius catchmentareas from the central business districts ofBerlin, Istanbul, Amsterdam, Rome, Barcelona,London, Paris, and Moscow. It is trained usingmorphological parameters to achieve high-accuracy land use classification which is latertested across multiple cities to identify howeffective the typological patterns of the buildingfabric condition land use patterns. The AI modelnot only predicts land use but also uncoverstypological patterns at an urban scale, offering acomparative and data-driven approach to urbanmorphology. By evaluating the accuracy of AI-generated predictions, this study contributesto the advancement of computational methodsin spatial analysis. Ultimately, this researchenhances our understanding of what makes anurban tissue unique or similar across differentcities, bridging AI-driven analysis with the long-standing theories of urban morphology.

URI

https://hdl.handle.net/11511/118064

Collections

Department of City and Regional Planning, Conference / Seminar