A 'bottom-up' hierarchical clustering algorithm that treats each data point as its own cluster and recursively merges the closest pairs until a single cluster (or a desired number) remains. The merger criteria are defined by 'linkage' methods: 'Single Linkage' uses the closest distance between points in two clusters, 'Complete Linkage' uses the farthest distance, and 'Ward's Method' minimizes the variance within the merged cluster. The result is typically visualized as a dendrogram, a tree-like diagram showing the order of merges.
Ward's method was introduced by Joe H. Ward Jr. in 1963.
Widely used in bioinformatics (phylogenetics), taxonomy, and image segmentation where data naturally forms a hierarchy.