IClusterMetric<T>
Interfaces · AiDotNet.Clustering.Evaluation
Interface for cluster evaluation metrics.
Known implementations: BCSS<T>, CalinskiHarabaszIndex<T>, ConnectivityIndex<T>, DaviesBouldinIndex<T>, DunnIndex<T>, FowlkesMallowsIndex<T>, SilhouetteScore<T>, VMeasure<T>, WCSS<T>
For Beginners
Cluster metrics answer “How good is this clustering?”
Two main types:
- Internal: Use data only (no “correct” answer needed)
- Silhouette Score: How similar are points to their own cluster?
- Davies-Bouldin: Are clusters compact and well-separated?
- Calinski-Harabasz: Ratio of between-cluster to within-cluster variance
- External: Compare to known labels
- Adjusted Rand Index: Agreement with ground truth
- Normalized Mutual Information: Information shared with ground truth
How It Works
Cluster metrics assess the quality of clustering results. They can be internal (using only the data) or external (comparing to ground truth labels).
Properties
| Property | Summary |
|---|---|
string Name { get; } |
Gets the name of this metric. |
Methods
| Method | Summary |
|---|---|
double Compute(Matrix<T> data, Vector<T> labels) |
Computes the metric value. |