Empirical Study of Partitions Similarity Measures

Alfalah, Abdlekrim; Ouarbya, Lahcen and Howroyd, John. 2021. 'Empirical Study of Partitions Similarity Measures'. In: ICCAE 2021: International Conference on Clustering Algorithms and Evaluation. Dubai, United Arab Emirates 20-21 December 2021. [Conference or Workshop Item]

Paper.pdf - Accepted Version
Available under License Creative Commons Attribution.

Download (636kB) | Preview

Abstract or Description

This paper compares four existing distances and similarity measures between partitions. The partition measures considered in this paper are the Rand Index (RI), the Adjusted Rand Index (ARI), the Variation of Information (VI) and finally, the Normalised Variation of Information (NVI). This work investigates the ability of these partition measures to capture three predefined intuitions: the variation within randomly generated partitions, the sensitivity to small perturbations and finally the independence from the dataset scale . It has been shown that the Adjusted Rand Index (ARI) performed well overall, regarding these three intuitions.

Item Type:

Conference or Workshop Item (Paper)


Clustering, comparing partitions, similarity measure, partition distance, partition metric, similarity between partitions, clustering comparison

Related URLs:

Departments, Centres and Research Units:



15 December 2021Accepted
20 December 2021Published

Event Location:

Dubai, United Arab Emirates

Date range:

20-21 December 2021

Item ID:


Date Deposited:

21 Dec 2021 15:18

Last Modified:

22 Dec 2021 03:17



View statistics for this item...

Edit Record Edit Record (login required)