Ward's Hierarchical Clustering Method: Clustering Criterion and Agglomerative Algorithm

摘要

The Ward error sum of squares hierarchical clustering method has been verywidely used since its first description by Ward in a 1963 publication. It hasalso been generalized in various ways. However there are differentinterpretations in the literature and there are different implementations ofthe Ward agglomerative algorithm in commonly used software systems, includingdiffering expressions of the agglomerative criterion. Our survey work and casestudies will be useful for all those involved in developing software for dataanalysis using Ward's hierarchical clustering method.