The most popular and extensively used method for analysing social network data is hierarchical clustering. Nodes are compared to one another using this method based on their similarity. Larger groups are formed by combining groups of nodes that are comparable in some way.
Hierarchical clustering can be divided into two types: divisive (top-down) and agglomerative (bottom-up) (bottom-up).
Another unsupervised learning approach, hierarchical clustering, is used to group together unlabeled data points with comparable features.
Divisive and Agglomerative hierarchical clustering are the two types of hierarchical clustering. We assign all of the observations to a single cluster in the divisive or top-down clustering technique, and then partition the cluster into two least similar clusters.
A hierarchical model is a data analysis structure in which the data is structured into a tree-like structure or one that uses multilevel (hierarchical) modelling. The first is concerned with both a theoretical framework and the placement of individual items under categories that may or may not be related.