Cluster labels in hierarchical clustering

Unsupervised Learning in Python

Benjamin Wilson

Director of Research at lateral.io

Cluster labels in hierarchical clustering

Not only a visualization tool!
Cluster labels at any intermediate stage can be recovered
For use in e.g. cross-tabulations

Eurovision hierarchical clustering

Intermediate clusterings & height on dendrogram

E.g. at height 15:
- Bulgaria, Cyprus, Greece are one cluster
- Russia and Moldova are another
- Armenia in a cluster on its own

Zoomed in cluster with horizontal line at height 15

Dendrograms show cluster distances

Height on dendrogram = distance between merging clusters
E.g. clusters with only Cyprus and Greece had distance approx. 6

Zoomed in cluster with Cyprus/Greece cluster highlighted

Dendrograms show cluster distances

Height on dendrogram = distance between merging clusters
E.g. clusters with only Cyprus and Greece had distance approx. 6
This new cluster distance approx. 12 from cluster with only Bulgaria

Zoomed in cluster with Cyprus/Greece cluster and Cyprus/Greece/Bulgaria cluster highlighted

Intermediate clusterings & height on dendrogram

Height on dendrogram specifies max. distance between merging clusters
Don't merge clusters further apart than this (e.g. 15)

Zoomed in cluster with horizontal line at height 15

Distance between clusters

Defined by a "linkage method"
In "complete" linkage: distance between clusters is max. distance between their samples
Specified via method parameter, e.g. linkage(samples, method="complete")
Different linkage method, different hierarchical clustering!

Extracting cluster labels

Use the fcluster() function
Returns a NumPy array of cluster labels

Zoomed in cluster with horizontal line at height 15

Extracting cluster labels using fcluster

from scipy.cluster.hierarchy import linkage
mergings = linkage(samples, method='complete')
from scipy.cluster.hierarchy import fcluster

labels = fcluster(mergings, 15, criterion='distance')
print(labels)

[ 9  8 11 20  2  1 17 14 ... ]

Aligning cluster labels with country names

Given a list of strings country_names:

import pandas as pd
pairs = pd.DataFrame({'labels': labels, 'countries': country_names})
print(pairs.sort_values('labels'))

               countries  labels
5                Belarus       1
40               Ukraine       1
...
36                 Spain       5
8               Bulgaria       6
19                Greece       6
10                Cyprus       6
28               Moldova       7
...

Let's practice!

Unsupervised Learning in Python