Weighted Clustering Ensemble: A Review

TL;DR

This paper reviews weighted clustering ensemble methods, discussing weight types, approaches to determine weights, and applications, aiming to guide practitioners in selecting suitable techniques for complex data analysis.

Contribution

It provides a comprehensive overview and unifying framework for weighted clustering ensemble, highlighting different weighting strategies and their applications.

Findings

Various weight types are discussed and categorized.

Major approaches for determining weights are summarized.

Applications to complex data demonstrate the framework's utility.

Abstract

Clustering ensemble, or consensus clustering, has emerged as a powerful tool for improving both the robustness and the stability of results from individual clustering methods. Weighted clustering ensemble arises naturally from clustering ensemble. One of the arguments for weighted clustering ensemble is that elements (clusterings or clusters) in a clustering ensemble are of different quality, or that objects or features are of varying significance. However, it is not possible to directly apply the weighting mechanisms from classification (supervised) domain to clustering (unsupervised) domain, also because clustering is inherently an ill-posed problem. This paper provides an overview of weighted clustering ensemble by discussing different types of weights, major approaches to determining weight values, and applications of weighted clustering ensemble to complex data. The unifying framework presented in this paper will help clustering practitioners select the most appropriate weighting mechanisms for their own problems.