[source]

Class Purity

A ground-truth clustering metric that measures the mean ratio of samples in a class that are also members of the class' dominant cluster. A clustering is said to be complete when all the samples in a class are contained in a single cluster.

\[ {\displaystyle Class\,Purity = {\frac {1}{m}}\sum _{j=1}^{m}{\frac {\max _{i}n_{ij}}{n_{j}}}} \]

Note

Since this metric monotonically improves as the number of target clusters decreases, it should not be used as a metric to guide hyper-parameter tuning.

Estimator Compatibility: Clusterer

Score Range: 0 to 1

Parameters

This metric does not have any parameters.

Example

use Rubix\ML\CrossValidation\Metrics\ClassPurity;

$metric = new ClassPurity();

Unlike Completeness, this metric does not use conditional entropy and tends to give more lenient scores on mixed assignments. See V Measure for the balanced entropy-based alternative.