The agglomerative hierarchical method, commonly referred to as hierarchical clustering, is a clustering approach that organizes data points into a hierarchical structure based on their similarity. This technique was introduced by Lance and Williams in 1967.
The process begins by treating each data point as its own cluster. Then, in an iterative manner, the two most similar clusters are merged until all points are combined into a single cluster. The similarity between clusters is determined using various distance metrics, such as Euclidean or Manhattan distance.
A key aspect of the algorithm is the linkage criterion, which defines how the distance between two clusters is calculated. This criterion is typically derived from the distances between individual data points within each cluster. Common linkage methods include single linkage, complete linkage, and average linkage.