Modeling the Economic Value of Real Estate under the Influence of Urban Infrastructure Development Using Explainable Artificial Intelligence

Modeling the Economic Value of Real Estate under the Influence of Urban Infrastructure Development Using Explainable Artificial Intelligence

Abed Samadi1

1) M.Sc. in Urban Planning, Tabriz Metropolitan Municipality, Tabriz, Iran Email:

IOI Code : ABZE-ABHGCC
Publication : 5th International Congress on Civil Engineering, Architecture, Building Materials and Environment(caucong.com)
Abstract :
    Accurate valuation of urban real estate is a fundamental component of efficient land management, taxation policy, and sustainable urban development. Urban green infrastructure, as one of the major determinants of property values, plays a key role in shaping land and real estate markets. However, traditional valuation methods often lack sufficient accuracy and transparency due to their limited ability to capture nonlinear relationships and interaction effects among influential factors. This study aims to develop a framework for modeling the economic value of urban real estate in response to green infrastructure development using Explainable Artificial Intelligence (XAI). The research adopts a descriptive–analytical approach and integrates Hedonic Pricing Theory with machine learning algorithms to analyze the determinants of property values across different spatial scales. The findings indicate that the Random Forest algorithm achieved the best predictive performance, with a coefficient of determination (R²) exceeding 0.83. Explainability analysis using Shapley values further reveals that proximity to green spaces within a 15-minute walking accessibility threshold has a greater influence on property values. Moreover, significant nonlinear and threshold effects were identified in the relationship between green space accessibility and urban property values.

Keywords : urban real estate valuation green infrastructure machine learning explainable artificial intelligence (xai) shapley value analysis