Back to metrics

ARI

C Clustering

Adjusted Rand Index

Direction

Higher is Better

Value Range

[-1, 1]

Category

Clustering

Agreement with reference labels

Description

Similarity between predicted and true clusters adjusted for chance

Mathematical Formula

ARI = (RI - E[RI]) / (max(RI) - E[RI])

Interpretation Guide

Range [-1, 1]. 1 = identical clusters, 0 = random clusters, -1 = opposite clusters. More robust to label permutations than NMI.

Related Metrics

Other metrics in the Clustering category

NMI

↑

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

ASW

↑

Average silhouette coefficient measuring cluster cohesion and separation

DAV

↓

Average similarity between each cluster and its most similar neighboring cluster