liveBook · Manning

Purity

As explained in Timeless Algorithms, The Foundational Ideas

Purity measures the extent to which each cluster is dominated by a single known category.

Purity describes how strongly each cluster is dominated by one known category. It evaluates whether cluster membership reveals meaningful labels, complementing geometric checks of how observations are arranged. section 5.3.4

Pure fruit clusters

If 12 fruits form three clusters containing only apples, only bananas, and only pears, respectively, each cluster is dominated by one category. Knowing a fruit’s cluster then determines its category. section 5.3.4

How Timeless Algorithms, The Foundational Ideas covers it

How deep it goes
Defines purity qualitatively as the dominance of one known category within each cluster.
Examples
Uses three fruit clusters, each containing one category, to illustrate complete alignment between clusters and labels.
What it assumes
Assumes known category labels are available for comparison.
What you can do after
Evaluate whether cluster assignments concentrate observations with the same known category.
How it connects
Connects purity with cluster entropy and normalized mutual information, distinguishing label-based evaluation from geometric evaluation.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Timeless Algorithms, The Foundational Ideas. The section links point to the parts of the book it draws on.