Back to metrics

DAV

🎯 Clustering & Cell Type Discovery

Davies-Bouldin Index

Direction

Lower is Better

Value Range

[0, ∞] arbitrary

Category

Clustering & Cell Type Discovery

Supervised metrics comparing predicted clusters to ground truth labels

Description

Average similarity between each cluster and its most similar neighboring cluster

Mathematical Formula

DB = (1/k) * Σ max(r_ij), where r_ij = (s_i + s_j) / d(c_i, c_j)

Interpretation Guide

Lower is better. 0 = ideal separation. Biased toward globular clusters. Computationally cheaper than silhouette.

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