Introduction to Clustering with Hierarchial Clustering

Table of contents

1. What is Hierarchial Clustering?

2. What is a Dendrogram?

1. What is Hierarchial Clustering?

Hierarchical clustering is a clustering algorithm that doesn’t require specifying the number of expected clusters beforehand. It measures the relationships between all observations in the dataset based on their similarity. The results are visualized in a tree-like structure called a dendrogram.There are two approaches that the algorithm takes to estimate the clusters:

1. Agglomerative: Every observation is considered to be a seperate cluster and we iteratively keep merging the clusters based on their similarity from each other till all the observations are a part of one cluster.

2. Divisive: All observations are considered to be a part of the same cluster and we iteratively keep removing observations based on their dissimilatiry from the other observations till each observation forms a seperate cluster.

Continue to the next session on DBSCAN by clicking here.

🧑🏽‍💻 Thank you for reading this document. If you have any feedback or want to make an addition to the information, you can email me at amandeepsinghkhanna@gmail.com.