The silhouette coefficient is a metric that doesn't need to know the labeling of the dataset. It gives an idea of the separation between clusters.
It is composed of two different elements:
- The mean distance between a sample and all other points in the same class (a)
- The mean distance between a sample and all other points in the nearest cluster (b)
The formula for this coefficient s is defined as follows:
The silhouette coefficient is only defined if the number of classes is at least two, and the coefficient for a whole sample set is the mean of the coefficient for all samples.
