The proliferation of spatial sources of “big data” – produced both by the private and public sector – in recent years has profoundly shaped the methodological and conceptual practice of geography and the social sciences more broadly. The availability of data at escalating volumes and velocities has led to the increasing adoption of newer analysis and visualization methods, including machine learning models that are not constrained by some of the traditional statistical assumptions of linear methods. These ongoing shifts in spatial data analytics have been perfectly illustrated by the experience of the COVID-19 pandemic, where data is required to be instantly available, structured, and provided via APIs.
In the context of this emerging data economy it is important for geographers 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. This is particularly important for economic geographers who are often most interested in the size and significance of explanatory relationships rather than in predicting outcomes.
In this special session we aim to explore the potentials of new data sources and related methods for economic geography. Potential topics include: