Groups - clusters

Observations of reality often have similarities, but just enough uniqueness to make them distinct. For example, when recording weather we typically record temperature, barometric pressure, etc. When displaying our data, we might notice that some observations are nearer to others sometimes forming noticeable groups or patterns.

Data showing clustered points. Original author:  Jake Vanderplas
Data showing clustered points. Original source:  Jake Vanderplas - Clustering: K-Means In-Depth

When observations appear in groups, we often call those groups 'clusters'. Clustering is process of dividing a set of observations into different clusters or groups on the basis of attributes.

Data showing clustered points.  Original source:  Jake Vanderplas
Data showing clustered points. Original source: Jake Vanderplas - Clustering: K-Means In-Depth
 PreviousNext