Skip to main content

Table 4 The best and worst values and the range of each metric

From: A set theory based similarity measure for text clustering and classification

Name

Best value

Worst value

Range

Accuracy/purity

1

0

0–1

Completeness

1

0

0–1

Rand Index

1.0

− 1.0

− 1.0–1.0

Calinski-Harabasz index

(Lower value is better)

–

–

Davies-Bouldin index

0

_

0–1