# Comparing high dimensional partitions, with the Coclustering Adjusted Rand Index

## TL;DR

This paper introduces the Co-clustering Adjusted Rand Index (CARI) for measuring agreement between co-clustering partitions, along with improvements to existing criteria, validated through experiments on simulated and real data.

## Contribution

It proposes a new co-clustering agreement measure, CARI, and enhances existing criteria like Classification Error and Normalized Mutual Information for better co-clustering evaluation.

## Key findings

- CARI effectively measures co-clustering agreement.
- Improved criteria show better properties in experiments.
- Experimental results validate the proposed measures.

## Abstract

We consider the simultaneous clustering of rows and columns of a matrix and more particularly the ability to measure the agreement between two co-clustering partitions. The new criterion we developed is based on the Adjusted Rand Index and is called the Co-clustering Adjusted Rand Index named CARI. We also suggest new improvements to existing criteria such as the Classification Error which counts the proportion of misclassified cells and the Extended Normalized Mutual Information criterion which is a generalization of the criterion based on mutual information in the case of classic classifications. We study these criteria with regard to some desired properties deriving from the co-clustering context. Experiments on simulated and real observed data are proposed to compare the behavior of these criteria.