The Potential of Machine Learning and Multi-Criteria Decision-Making for Data-Driven Land Policies in Türkiye

Abstract. Land management policies are shaped through diverse development processes and implemented via instruments such as legal regulations, incentives, and land development tools. Especially in urban areas, the complexity of social, economic, and environmental dynamics requires policy-making that is adaptive, predictive, and data-driven. Global agendas, including the UN’s Sustainable Development Goals, highlight issues such as equitable land access, resource efficiency, and informal settlement upgrading. In this context, the success of policy-making depends on effectively managing uncertainty and utilizing both past experiences and real-time data. As decisions grow more complex and data volumes increase, advanced analytical tools become essential. Machine learning (ML) algorithms enable pattern recognition and prediction from large datasets, while Multi-Criteria Decision-Making (MCDM) methods offer structured evaluation of alternatives based on multiple criteria and stakeholder inputs. The integration of ML and MCDM provides a comprehensive framework to support dynamic, informed, and sustainable land policy development and implementation. This study aims to examine the potential application of machine learning and MCDM methods within the context of land management policies in Türkiye. This study suggests that ML and MCDM have high potential to support improvements in Türkiye's land management policies. The potential of these tools shows that policymakers can benefit from them to make more informed, data-driven decisions, ensure more efficient and equitable land management, and ultimately contribute to sustainable urban and rural development.