Silhouette Coefficient
A clustering quality evaluation metric that jointly measures intra-cluster compactness and inter-cluster separation, with values in the range [-1, 1].
Timeline (last 90 days)
In the experiments, different consonant categories showed consistent geometric separation in the two-dimensional space, with a silhouette coefficient exceeding 0.5.
The silhouette coefficient is a metric for evaluating clustering quality, ranging from -1 to 1, with values closer to 1 indicating tighter intra-cluster compactness and greater inter-cluster separation.
The silhouette coefficient computes, for each sample, a (the average distance to other points in the same cluster) and b (the average distance to all points in the nearest other cluster), and takes (b-a)/max(a,b), which ranges from [-1,1], with values closer to 1 indicating compact clusters and well-separated clusters.
All Facts (3)
In the experiments, different consonant categories showed consistent geometric separation in the two-dimensional space, with a silhouette coefficient exceeding 0.5.
50%UnverifiedThe silhouette coefficient is a metric for evaluating clustering quality, ranging from -1 to 1, with values closer to 1 indicating tighter intra-cluster compactness and greater inter-cluster separation.
50%UnverifiedThe silhouette coefficient computes, for each sample, a (the average distance to other points in the same cluster) and b (the average distance to all points in the nearest other cluster), and takes (b-a)/max(a,b), which ranges from [-1,1], with values closer to 1 indicating compact clusters and well-separated clusters.
50%