Tag Archives: Silhouette coefficient

Silhouette coefficient for finding optimal number of clusters

Silhouette coefficient is another method to determine the optimal number of clusters. Here I introduced c-index earlier. The silhouette coefficient of a data measures how well data are assigned to its own cluster and how far they are from other clusters. A silhouette close to 1 means the data points are in an appropriate cluster and a silhouette coefficient close to −1 implies out data is in the wrong cluster. The following is python code for computing the coefficient and plotting number fo clusters vs Silhouette coefficient.