1.1 Overview of Volume 2

Volume 2 is organized into four parts:

  • Dimension reduction

  • Classic clustering

  • Spatial clustering

  • Assessment

The first part reviews classic dimension reduction techniques, divided into three chapters, devoted to principal components, multidimensional scaling and stochastic neighbor embedding. In addition to a discussion of the classic properties, specific attention is paid to spatializing these techniques, i.e., bringing out interesting spatial aspects of the results.

Part II covers classic clustering methods, in contrast to spatially constrained clusters, which are the topic of Part III. Four chapters deal with, respectively, hierarchical clustering methods, partitioning clustering methods (K-Means), advanced methods (K-Medians and K-Medoids), and spectral clustering.

The chapters in Part III deal with methods to include an explicit spatial constraint of contiguity into the clustering routine. The first chapter outlines techniques to spatialize classic clustering methods, which involve soft spatial constraints. These techniques do no not guarantee a spatially compact (contiguous) solution. In contrast, the methods discussed in the next two chapters impose hard spatial constraints. One chapter deals with hierarchical approaches (spatially constrained hierarchical clustering, SKATER and REDCAP), the other with partitioning methods (AZP and max-p).

Part IV deals with assessment and includes a final chapter outlining a range of approaches to validate the cluster results, both in terms of internal validity as in terms of external validity. It closes with some concluding remarks.

As before, in addition to the material covered in this volume, the GeoDaCenter Github site (https://geodacenter.github.io) contains an extensive support infrastructure. This includes detailed documentation and illustrations, as well as a large collection of sample data sets, cookbook examples and links to a YouTube channel containing lectures and tutorials. Specific software support is provided by means of a list of frequently asked questions and answers to common technical questions, as well as by the community through the Google Groups Openspace list.