Hands-On Machine Learning with C++

Hierarchical clustering algorithms

Among the algorithms of hierarchical clustering, there are two main types: bottom-up- and top-down-based algorithms. Top-down algorithms work on the next principle: at the beginning, all objects are placed in one cluster, which is then divided into smaller and smaller clusters. Bottom-up algorithms are more common than top-down ones. They place each object in a separate cluster at the beginning of the work, and then merge clusters into larger ones until all the objects in the dataset are contained in the one cluster, building a system of nested partitions. The results of such algorithms are usually presented in tree form, called a dendrogram. A classic example of such a tree is the tree of life, which describes ...