Integration of AI, Spatial Data, and GIS in Planning: Spatial Application Based on Machine Learning and Deep Learning

The study focuses on the integration of data, deep learning (DL) models, and machine learning (ML) algorithms with geographical information systems (GIS) within the field of spatial planning. An original contribution is provided by addressing the integration of DL and ML into GIS in terms of their advantages, limitations, encountered challenges, and potential directions for development within the context of spatial data and model validation processes. In this context, the objective is to identify the developmental trajectory, challenges, and potentials of spatial studies based on the integration of DL, ML, and GIS. To achieve this aim, 91 research articles published in high-impact journals indexed in the Web of Science (WoS) database were analyzed. The selected studies were evaluated under five main categories: spatial and temporal distribution, applications of DL and ML methods, thematic approaches, employed GIS tools, and data-model validation processes. The findings suggest that artificial intelligence technologies have the potential to serve as significant tools in spatial planning, although the current developmental stage remains in its early phases. While ML algorithms are widely applied across the reviewed studies, the application scope of DL models has expanded in recent years due to the increasing availability of large datasets. Spatial applications predominantly concentrate on land use, natural hazard assessments, environmental issues, and climate-related themes, particularly supported by the extensive use of remote sensing techniques. However, due to the limited accessibility of spatial data in rural areas, the majority of applied studies have been oriented toward urban centers, revealing a noticeable gap in research focusing on rural contexts. Furthermore, studies that implement AI and planning integration in practice demonstrate that the use of spatial data and the necessity of model validation constitute critical requirements. This study may offer guidance for future research by supporting the implementation of applications across diverse thematic domains involving the integration of artificial intelligence, planning, and GIS within spatially oriented processes.