Journal of Geographical Systems - Call for Papers: Special Issue on “Spatial Machine Learning: Perspectives, Methods, and Applications”

Guest Editors:

Kevin Credit (Lead Guest Editor)
Assistant Professor
National Centre for Geocomputation
Maynooth University
kevin.credit@mu.ie (this opens in a new tab)

Matthew Lehnert
Lecturer of Business & Statistics
School of Business Administration
Al Akhawayn University
m.lehnert@aui.ma (this opens in a new tab)
 

Aims and Scope:

New predictive machine learning methods and techniques have become more widely used, and the emerging field of “data science” is increasingly engaging to judge these methods in many application domains, including the social sciences. It is therefore important for geographers, regional scientists, and urban planners to better understand how these methods operate, how they compare to traditional approaches, and how they use spatial information. In particular, the integration of spatial data into these models and the creation of spatially-explicit machine learning models are very important, but relatively undeveloped areas of research. At the same time, the efficacy of these powerful predictive methods (such as random forest) for explanatory research questions remains inadequately understood, underscoring the need to further open the “black box” of machine learning algorithms. This is particularly important for social scientists who are often most interested in the size and significance of explanatory relationships, rather than predicting outcomes.

The purpose of this special issue is to spur a wide-ranging conversation about the usefulness and applicability of machine learning methods in geography, regional science and related disciplines. It aims to serve as a showcase for work that develops new spatially-explicit machine learning methods, or uses these techniques in innovative applications.
We welcome papers from across the disciplinary spectrum that employ new machine learning techniques, or discuss the development or approaches of “data science” as it relates to geographic data and analysis, and spatial ways of thinking more broadly, including on topics such as:

• Integration of spatial data into predictive machine learning methods and techniques such as random forest
• Structured comparisons between “newer” machine learning models and traditional (spatial) econometric approaches
• Methods for optimizing spatial pattern prediction or the development of new indicators of spatial association
• Development or use of explanatory machine learning models, including causal random forests (CRF)
• Use of new visualization methods for non-linear relationships in machine learning models, such as partial dependence (PD) and accumulated local effects (ALE) plots
• Development or use of GeoAI and deep learning models for spatial data science applications

Tentative schedule:

All submissions received: August 31, 2022
All final decisions made:  July 2023

Submit a manuscript (this opens in a new tab):
The system is set up to accept submissions to the special issue “Spatial Machine Learning: Perspectives, Methods, and Applications”. Please select the submission type “S.I.: Spatial Machine Learning” from the drop-down menu.