Journal of Digital Landscape Architecture

Seite drucken

Artificial Intelligence and Learning from Nature in Landscape Architecture: An Innovative Approach that Shapes the Analysis & Design Process

Today, artificial intelligence (AI), as well as making our lives easier, has become an essential component in professional fields, including landscape architecture (LA). It has been integrated into architecture & design software by being developed by data scientists, as well as designers and planners. This development minimizes the time and energy to be spent in the process of comprehensive site analysis. In this study, rather than relying solely on existing architectural software, the aim is to explore the capabilities of ChatGPT as a customizable AI tool for site analysis by integrating domain-specific data and coding inputs. In the near future, by development of these tools by designers and planners, problem detection can be done more clearly, and more qualified designs can be made in a short time. In the process of training AI, Kültürpark in Izmir was chosen as the study area as it has been subjected to various physical and functional changes since the beginning of the 20th century. This research contributes to the development of AI-enabled decision support systems in LA and highlights the importance of human-in-the-loop (HITL) approach, where human intervention validates AI outputs for optimal results.

Autor / Author: Er, Buse Ezgi; Bozkurt, Melih
Institution / Institution: Istanbul Technical University, Istanbul/Turkey; Istanbul Technical University, Istanbul/Turkey
Seitenzahl / Pages: 9
Sprache / Language: Englisch
Veröffentlichung / Publication: JoDLA – Journal of Digital Landscape Architecture, 10-2025
Tagung / Conference: Digital Landscape Architecture 2025 – Collaboration
Veranstaltungsort, -datum / Venue, Date: Dessau Campus of Anhalt University, Germany 04-06-25 - 07-06-25
Schlüsselwörter (de):
Keywords (en): Artificial intelligence, ChatGPT, human-in-the-loop, landscape architecture, Python
Paper review type: Full Paper Review
DOI: doi:10.14627/537754038
8367 -