Hierarchical clustering is connectivity-based clustering. It assumes that the clusters are connected, or in another word, linked. For example, we can classify animals and plants based on this assumption. We have all developed from something common. This makes it possible for us to assume that every observation is its own cluster on one hand and, on the other, all observations are in one and the same group. This also forms the basis for two approaches to hierarchical clustering algorithms, agglomerative and divisive:
Agglomerative clustering starts out with each point in its own cluster and then merges the two clusters with the lowest dissimilarity, that is, the bottom-up approach
Divisive clustering is, as the name suggests, a top-down approach where we start out with one single cluster that is divided into smaller and smaller clusters
In contrast to k-means, it gives us a way to identify the clusters without initial guesses of the number of clusters or cluster...