Back to metrics

CAL

🎯 Clustering & Cell Type Discovery

Calinski-Harabasz Index

Direction

Higher is Better

Value Range

[0, ∞]

Category

Clustering & Cell Type Discovery

Supervised metrics comparing predicted clusters to ground truth labels

Description

Ratio of between-cluster to within-cluster dispersion

Mathematical Formula

CH = (B / (k-1)) / (W / (n-k))

Interpretation Guide

Higher is better. Assumes convex, isotropic clusters. Fast to compute. Biased toward many small clusters.

Related Metrics

Other metrics in the Clustering & Cell Type Discovery category

NMI

↑

Measures the mutual information between predicted clusters and true labels, normalized by the entropy of the two distributions

ARI

↑

Similarity between predicted and true clusters adjusted for chance

ASW

↑

Average silhouette coefficient measuring cluster cohesion and separation