Ali Shams
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ali Shams has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Hossein Tahmasebi Moghaddam
Department of Geography, Faculty of Human Sciences, University of Zanjan, Zanjan, Iran
1 shared publication - Khalil Gholamnia
Department of Applied Geoinformatics and Cartography, Charles University, Prague 2, Czech Republic
1 shared publication - Omid Ghorbanzadeh
Department of Geoinformatics—Z_GIS, University of Salzburg, Salzburg, Austria
1 shared publication Shadali Tohidloo
Department of Extension, Communication and Rural Development, Faculty of Agriculture University of Zanjan Zanjan Iran
1 shared publication- Vahide Mollakarami
Department of Extension, Communication and Rural Development, Faculty of Agriculture University of Zanjan Zanjan Iran
1 shared publication
Latest Publications
- Open Access
- 27 October 2026
- Published by Wiley in Transactions in GIS
- Vol. 30 (7)
Sustainable Land Suitability Assessment for Garden City Development in Zanjan's Urban Hinterland Using Machine Learning
In response to rapid urban expansion, population growth, and the need for sustainable land use, urban agriculture presents a viable strategy for enhancing food security and ecological balance in urban planning. This study aims to assess land suitability for garden city development in the urban hinterland zone of Zanjan. A hybrid framework integrating the Analytic Hierarchy Process (AHP) with two machine learning (ML) methods, Support Vector Machine (SVM) and Extreme Gradient Boosting (XGBoost), was employed. A set of geospatial and environmental criteria, including slope, aspect, elevation, soil erosion, climatic variables (precipitation and temperature), and proximity to infrastructure and natural features, were analyzed to identify ecologically and socially suitable locations for garden city development. The standardized criteria were used as predictor variables, while expert‐informed suitability scores derived from the AHP framework served as reference values for model development. The results showed that the SVM and XGBoost models successfully learned the spatial suitability patterns represented by the expert‐informed AHP framework, achieving coefficients of determination of R 2 = 0.988 and R 2 = 0.977, respectively. These values indicate a high level of agreement between the ML predictions and the expert‐informed suitability patterns rather than independent predictive accuracy. The findings also revealed that distance from faults, access to urban infrastructure, and land slope were the most influential factors in determining suitable areas. The proposed hybrid framework demonstrates how expert knowledge and data‐driven learning can be integrated to support transparent and comprehensive land suitability assessment for garden city development in peri‐urban environments.
- 27 October 2026
- Published by Wiley in Transactions in GIS
- Vol. 30 (7)
Open AccessSustainable Land Suitability Assessment for Garden City Development in Zanjan's Urban Hinterland Using Machine Learning
In response to rapid urban expansion, population growth, and the need for sustainable land use, urban agriculture presents a viable strategy for enhancing food security and ecological balance in urban planning. This study aims to assess land suitability for garden city development in the urban hinterland zone of Zanjan. A hybrid framework integrating the Analytic Hierarchy Process (AHP) with two machine learning (ML) methods, Support Vector Machine (SVM) and Extreme Gradient Boosting (XGBoost), was employed. A set of geospatial and environmental criteria, including slope, aspect, elevation, soil erosion, climatic variables (precipitation and temperature), and proximity to infrastructure and natural features, were analyzed to identify ecologically and socially suitable locations for garden city development. The standardized criteria were used as predictor variables, while expert‐informed suitability scores derived from the AHP framework served as reference values for model development. The results showed that the SVM and XGBoost models successfully learned the spatial suitability patterns represented by the expert‐informed AHP framework, achieving coefficients of determination of R 2 = 0.988 and R 2 = 0.977, respectively. These values indicate a high level of agreement between the ML predictions and the expert‐informed suitability patterns rather than independent predictive accuracy. The findings also revealed that distance from faults, access to urban infrastructure, and land slope were the most influential factors in determining suitable areas. The proposed hybrid framework demonstrates how expert knowledge and data‐driven learning can be integrated to support transparent and comprehensive land suitability assessment for garden city development in peri‐urban environments.