Unsupervised ensemble minority clustering

Cluster analysis lies at the core of most unsupervised learning tasks. However, the majority of clustering algorithms depend on the all-in assumption, in which all objects belong to some cluster, and perform poorly on minority clustering tasks, in which a small fraction of signal data stands against a majority of noise. The approaches proposed so far for minority clustering are supervised: they require the number and distribution of the foreground and background clusters. In supervised learning and all-in clustering, combination methods have been successfully applied to obtain distribution-free learners, even from the output of weak individual algorithms. In this work, we propose a novel ensemble minority clustering algorithm, Ewocs , suitable for weak clustering combination. Its properties have been theoretically proved under a loose set of constraints. We also propose a number of weak clustering algorithms, and an unsupervised procedure to determine the scaling parameters for Gaussian kernels used within the task. We have implemented a number of approaches built from the proposed components, and evaluated them on a collection of datasets. The results show how approaches based on Ewocs are competitive with respect to—and even outperform—other minority clustering approaches in the state of the art.