Understanding the output of clustering
Different approaches may be used to understand the output of clustering including:
1. Building a decision tree (DT) with the Cluster Label as the Target variable, and using the associated rules to conveniently
describe each cluster as well as explain how to assign new records to the correct cluster (e.g. [42]).
2. Examining the distribution of variable values from cluster to cluster. Typically this involves the domain expert(s) doing
comparison of cluster means for the relevant variables