SELECTIVELY LEARNING CLUSTERS IN MULTI-EAC

André Lourenço, Ana Fred

2010

Abstract

The Multiple-Criteria Evidence Accumulation Clustering (Multi-EAC) method, is a clustering ensemble approach with an integrated cluster stability criterion used to selectively learn the similarity from a collection of different clustering algorithms. In this work we analyze the original Multi-EAC criterion in the context of the classical relative validation criteria, and propose alternative cluster validation indices for the selection of clusters based on pairwise similarities. Taking several clustering ensemble construction strategies as context, we compare the adequacy of each criterion and provide guidelines for its application. Experimental results on benchmark data sets show the proposed concepts.

Paper Citation

in Harvard Style

Lourenço A. and Fred A. (2010). SELECTIVELY LEARNING CLUSTERS IN MULTI-EAC . In Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010) ISBN 978-989-8425-28-7, pages 491-499. DOI: 10.5220/0003099904910499

in Bibtex Style

@conference{kdir10,
author={André Lourenço and Ana Fred},
title={SELECTIVELY LEARNING CLUSTERS IN MULTI-EAC},
booktitle={Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010)},
year={2010},
pages={491-499},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003099904910499},
isbn={978-989-8425-28-7},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010)
TI - SELECTIVELY LEARNING CLUSTERS IN MULTI-EAC
SN - 978-989-8425-28-7
AU - Lourenço A.
AU - Fred A.
PY - 2010
SP - 491
EP - 499
DO - 10.5220/0003099904910499