Predicting industrial property prices with explainable artificial intelligence

Tris Kee & Winky K.O. Ho

Journal of Property Research2025https://doi.org/10.1080/09599916.2025.2550976article
ABDC B
Weight
0.50

Abstract

The industrial property market in Hong Kong is a dynamic and complex sector, marked by unique characteristics and atypical market behaviour. This study leverages the predictive power of Gradient Boosting Machines (GBM) to uncover the intricate relationships that drive property prices. Key features such as location, square footage, floor level and proximity to mass transit railway stations are analysed, with Shapley values providing a transparent and interpretable measure of each feature’s impact. Our findings reveal striking non-linear interactions among these features, vividly depicted through beeswarm plots showcasing wide SHAP value distributions that oscillate across the baseline. These insights illuminate the nuanced interplay between property attributes and their market valuation, offering a fresh perspective on the industrial property sector. By equipping stakeholders with actionable intelligence, this research empowers data-driven decision-making, fostering a deeper understanding of the forces shaping property prices in one of the world’s most dynamic real estate markets.

Open via your library →

Cite this paper

https://doi.org/https://doi.org/10.1080/09599916.2025.2550976

Or copy a formatted citation

@article{tris2025,
  title        = {{Predicting industrial property prices with explainable artificial intelligence}},
  author       = {Tris Kee & Winky K.O. Ho},
  journal      = {Journal of Property Research},
  year         = {2025},
  doi          = {https://doi.org/https://doi.org/10.1080/09599916.2025.2550976},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Predicting industrial property prices with explainable artificial intelligence

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.50

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.50 × 0.4 = 0.20
M · momentum0.50 × 0.15 = 0.07
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.