<aside> 💡
Cluster analysis groups similar objects together while keeping dissimilar objects in different groups.
</aside>
Cluster analysis is an unsupervised learning method. It help us to identify natural patterns and structures within data, making it valuable for tasks like customer segmentation, image processing, and pattern recognition.
In the figure on the left, the data is un-clustered. The figure on the right indicates how the clustering algorithm has grouped the datapoints based on their similarity and difference to the other datapoints.
The core objective of clustering is to maximise both intra-cluster similarity (objects within the same cluster should be as similar as possible) and inter-cluster differences (objects in different clusters should be as distinct as possible).
Review each of the follow elements (in order):
To help consolidate your learning, review the revisions questions here:
Revision Questions - Cluster Analysis
Presentation as PDF