Machine Learning in the Design Decision-Making of Traditional Garden Space Renewal: A Case Study of the Classical Gardens of Jiangnan

A machine learning approach using conditional generative adversarial networks (CGAN) achieves high accuracy in generating traditional garden space schemes, with a 91.08% success rate in capturing key elements of Suzhou Gardens. This method offers an efficient alternative to human-experience-based design, enabling rapid generation and evaluation of multiple renewal plans for spaces ranging from 200 to 1000 square meters. The study demonstrates the potential of AI-driven tools to support decision-making in urban renewal projects by providing data-backed comparisons and feasibility analyses.

Yan, L., Zheng, L., Jia, X., Zhang, Y., & Chen, Y. (2025). Machine Learning in the Design Decision-Making of Traditional Garden Space Renewal: A Case Study of the Classical Gardens of Jiangnan. Buildings, 15(14), 2401. https://doi.org/10.3390/buildings15142401