Comparison among Methods for k Estimation in k-means

Murilo C. Naldi, André Fontana, Ricardo J.G.B. Campello

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

One of the most influential algorithms in data mining, k-means, is broadly used in practical tasks for its simplicity, computational efficiency and effectiveness in high dimensional problems. However, k-means has two major drawbacks, which are the need to choose the number of clusters, k, and the sensibility to the initial prototypes' position. In this work, systematic, evolutionary and order heuristics used to suppress these drawbacks are compared. 27 variants of 4 algorithmic approaches are used to partition 324 synthetic data sets and the obtained results are compared.

Original languageEnglish
Title of host publicationISDA 2009 - 9th International Conference on Intelligent Systems Design and Applications
PublisherIEEE
Publication date2009
Pages1006-1013
Article number5364434
ISBN (Print)9780769538723
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event9th International Conference on Intelligent Systems Design and Applications, ISDA 2009 - Pisa, Italy
Duration: 30. Nov 20092. Dec 2009

Conference

Conference9th International Conference on Intelligent Systems Design and Applications, ISDA 2009
Country/TerritoryItaly
CityPisa
Period30/11/200902/12/2009
SponsorInternational Fuzzy Systems Association, MIR Labs, Universidad de Granada, UGR, Universita de Pisa, University of Salerno
SeriesInternational Conference on Intelligent Systems Design and Applications, ISDA
ISSN2164-7143

Fingerprint

Dive into the research topics of 'Comparison among Methods for k Estimation in k-means'. Together they form a unique fingerprint.

Cite this