Hierarchical clustering

Hierarchical clustering, also known as hierarchical cluster analysis, is a clustering algorithm that groups similar objects into groups called clusters. Hierarchical clustering is often associated with heatmaps.

Strategies

Strategies for hierarchical clustering generally fall into two types:

• Divisive: This type works on the assumption that all the feature vectors form a single set and then hierarchically go on dividing this group into different sets.
• Agglomerative: The idea is to ensure nearby points end up in the same cluster. In this type, partitions are visualized using a tree structure called dendrogram.

Advantages vs disadvantages

Advantages

• Hierarchical clustering does not require the number of clusters to be specified.
• It is easy to implement.
• Hierarchical clustering produces a dendogram, which hlps with understanding the data.

Disadvantages

• The hierarchical algorithm can never do any previous steps throughout the algorithm
• The time complexity for the clustering can result in very long computation times in comparison with efficient algorithms such as K-means.
• If we have a large data set, it can become difficult to determine the correct number of clusters by the dendrogram.