Back to metrics

NMI

🎯 Clustering & Cell Type Discovery

Normalized Mutual Information

Direction

Higher is Better

Value Range

[0, 1]

Category

Clustering & Cell Type Discovery

Supervised metrics comparing predicted clusters to ground truth labels

Description

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

Mathematical Formula

NMI(Y, Ŷ) = I(Y; Ŷ) / √(H(Y) * H(Ŷ))

Interpretation Guide

Range [0, 1]. 1 = perfect clustering match with ground truth. Symmetric, accounts for chance clustering.

Related Metrics

Other metrics in the Clustering & Cell Type Discovery category