Weighted clustering ensemble: A review

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