Extracting silhouette information from clustering
Silhouette information is a measurement to validate a cluster of data. In the previous recipe, we mentioned that the measurement of a cluster involves the calculation of how closely the data is clustered within each cluster, and measures how far different clusters are apart from each other. The silhouette coefficient combines the measurement of the intracluster and intercluster distance. The output value typically ranges from 0 to 1; the closer to 1, the better the cluster is. In this recipe, we will introduce how to compute silhouette information.
In order to extract the silhouette information from a cluster, you need to have the previous recipe completed by generating the