What is the difference between agglomerative and divisive hierarchical clustering?

Data Science, AI and ML
#1
  • The agglomerative hierarchical clustering technique reads clusters from bottom to top, thus the programme always starts with the sub-component and then continues on to the parent; in contrast, divisive hierarchical clustering reads clusters from top to bottom, starting with the parent and then the child.
  • The agglomerative hierarchical technique is made up of items that each produce their own clusters, which are then joined together to form a larger cluster. It is a continuous merging process that will continue until all of the single clusters have been merged into a single large cluster that contains all of the items from the child clusters. Divisive clustering, on the other hand, divides the parent cluster into smaller clusters and continues to split until each cluster has a single object to represent.