Connectivity-based Clustering

Connectivity-based methods group data points together based on the proximity between the clusters.

Calculate proximity between the clusters using linkage criterion

## Hierarchical Clustering

How it works

  1. Make each data point a single-point cluster => that forms N clusters
  2. Take the two closest data points and make them one cluster => that forms N-1 clusters
  3. Take the two closest data clusters and make them one cluster => that forms N-2 clusters
  4. Repeat until there is only one cluster

Dendrograms

/assets/images/hierarchical-clustering-1.png|600

Dendrograms remain the memory of each step in hierarchical clustering.

example:
/assets/images/hierarchical-clustering-2.png|600