Pivotal seeding for K-means based on clustering ensembles

Autor: Leonardo Egidi, Roberta Pappadà, Francesco Pauli, Nicola Torelli
Přispěvatelé: Società Italiana di Statistica, Giuseppe Arbia, Stefano Peluso, Alessia Pini, Giulia Rivellini, Egidi, Leonardo, Pappada', Roberta, Pauli, Francesco, Torelli, Nicola
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Popis: Despite its large use, one major limitation of K-means algorithm is the impact of the initial seeding on the final partition. We propose a modified version, using the information contained in a co-association matrix obtained from clustering ensembles; such matrix is given as input for a set of pivotal methods, implemented in the pivmet R package, used to perform a pivot-based initialization step. Preliminary results concerning the comparison with the classical approach and other clustering methods are discussed.
Databáze: OpenAIRE