14th May 2020, 2pm

Roberta Pappadà (University of Trieste)

Title: Consensus clustering based on pivotal methods

Abstract: Despite its large use, one major limitation of K-means clustering algorithm is its sensitivity to the initial seeding used to produce the final partition. We propose a modified version of the classical approach, which exploits the information contained into a co-association matrix obtained from clustering ensembles. Our proposal is based on the identification of a set of data points–pivotal units–that are representative of the group they belong to. The presented approach can thus be viewed as a possible strategy to perform consensus clustering. The selection of pivotal units has been originally employed for solving the so-called label-switching problem in Bayesian estimation of finite mixture models. Different criteria for identifying the pivots are discussed and compared. We investigate the performance of the proposed algorithm via simulation experiments and the comparison with other consensus methods available in the literature.