Complete linkage clustering

Complete linkage clustering is a form of hierarchical clustering method in which the distance between two clusters is defined as the maximum distance between any two points in each of the two clusters. This method is sometimes referred to as the “farthest neighbor” or “maximum mating” method. Unlike other clustering methods, complete linkage clustering can be used to discover clusters of different shapes and sizes within a dataset. In this method, the distance between two clusters is determined by the two-point most dissimilar data points in each of the two clusters. The two-point most dissimilar data points are calculated by finding the greatest difference between any two points in each cluster. The cluster with the highest difference is selected as the one with the maximum distance and the two clusters are merged. The process is then repeated for all pairs of clusters until the desired number of clusters is obtained. Complete linkage clustering is a good choice for data with a high degree of variability and noise. It works well with outliers, as it is less affected by outliers than other clustering algorithms. It also works well with different shapes and sizes of clusters and can be effective with small sample sizes.