Description

A cluster is a collection of objects which are “similar” between them and are “dissimilar” to the objects belonging to other clusters. Cluster ensemble means collection of individual solutions. Relative Validity Indexes are used to evaluate the quality of the clusters. The goal of clustering is the discovery of a meaningful and consistent cluster structure for a given collection of data objects. The overproduction of clusters in base partitions as a method to enhance diversity in ensembles.