Developing geodemographic clusters

To begin your clustering exercise, it is helpful to talk about a few types of clustering algorithms that you’ll be leveraging within this section. Your first model will be developed using a K-means clustering algorithm. The K-means clustering algorithm aims to split your observations into a predefined number of clusters that minimizes within-cluster variance. Within-cluster variance measures the similarity of observations that are grouped together in the same cluster. Later on in this section, we’ll discuss how to develop clustering models using an agglomerative hierarchical clustering (AHC) algorithm. Agglomerative clustering begins with each observation in its own cluster and recursively merges pairs of clusters together based on a linkage metric. Similar to K-means, AHC aims to minimize within-cluster variance while maximizing between-cluster variance. There are many other types of clustering algorithms out there, such as density...