Summary
Ensemble clustering is a novel research field that extends to unsupervised learning the approach originally developed for classification and supervised learning problems. In particular ensemble clustering methods have been developed to improve the robustness and accuracy of clustering algorithms, as well as the ability to capture the structure of complex data. In many clustering applications an example may belong to multiple clusters, and the introduction of fuzzy set theory concepts can improve the level of flexibility needed to model the uncertainty underlying real data in several application domains. In this paper, we propose an unsupervised fuzzy ensemble clustering approach that permit to dispose both of the flexibility of the fuzzy sets and the robustness of the ensemble methods. Our algorithmic scheme can generate different ensemble clustering algorithms that allow to obtain the final consensus clustering both in crisp and fuzzy formats.