ABSTRACT
GIS and machine learning (ML) are important branches of data science. GIS facilitates the theorizing and deployment of techniques that allow the creation, manipulation, and display of digital data for the purpose of spatial problem solving. ML, on the other hand, incorporates techniques that help to make sense out of noisy data by unmasking hidden patterns in the data. This chapter focuses on how some ML methods can be combined with GIS principles and practices to develop small area classifications that are useful for policy makers. The discussions here cover an exhaustive list of processes and procedures. Not only does the chapter address the range of technical considerations and ML algorithms that underpin the development of area classifications, careful attention is also given to some rules of thumb for detecting natural clusters, sensitivity analysis, and the art and science of naming and describing classification outputs.
