Creating Compact Regions of Social Determinants of Health
This work addresses the lack of comparative analysis for regionalization algorithms on large-scale real-world data, which is incremental but important for applications in public health and geography.
This study compared state-of-the-art regionalization methods using real-world social determinants of health data with up to 1 million points, evaluating them with new geographic metrics and memory analysis, and found that the best method outperformed unconstrained K-Means clustering in separating health data in Virginia and Washington DC.
Regionalization is the act of breaking a dataset into contiguous homogeneous regions that are heterogeneous from each other. Many different algorithms exist for performing regionalization; however, using these algorithms on large real world data sets have only become feasible in terms of compute power in recent years. Very few studies have been done comparing different regionalization methods, and those that do lack analysis in memory, scalability, geographic metrics, and large-scale real-world applications. This study compares state-of-the-art regionalization methods, namely, Agglomerative Clustering, SKATER, REDCAP, AZP, and Max-P-Regions using real world social determinant of health (SDOH) data. The scale of real world SDOH data, up to 1 million data points in this study, not only compares the algorithms over different data sets but provides a stress test for each individual regionalization algorithm, most of which have never been run on such scales previously. We use several new geographic metrics to compare algorithms as well as perform a comparative memory analysis. The prevailing regionalization method is then compared with unconstrained K-Means clustering on their ability to separate real health data in Virginia and Washington DC.