K-means clustering with multiresolution peak detection

Guanshan Yu, Leen Kiat Soh, Alan Bond

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Scopus citations

Abstract

Clustering is a practical data mining approach of pattern detection. Because of the sensitivity of initial conditions, k-means clustering often suffers from low clustering performance. We present a procedure to refine initial conditions of k-means clustering by analyzing density distributions of a data set before estimating the number of clusters k necessary for the data set, as well as the positions of the initial centroids of the clusters. We demonstrate that this approach indeed improves the accuracy and performance of k-means clustering measured by average intra to interclustering error ratio. This method is applied to the virtual ecology project to design a virtual blue jay system.

Original languageEnglish (US)
Title of host publication2005 IEEE International Conference on Electro Information Technology
StatePublished - 2005
Event2005 IEEE International Conference on Electro Information Technology - Lincoln, NE, United States
Duration: May 22 2005May 25 2005

Publication series

Name2005 IEEE International Conference on Electro Information Technology
Volume2005

Conference

Conference2005 IEEE International Conference on Electro Information Technology
CountryUnited States
CityLincoln, NE
Period5/22/055/25/05

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'K-means clustering with multiresolution peak detection'. Together they form a unique fingerprint.

  • Cite this

    Yu, G., Soh, L. K., & Bond, A. (2005). K-means clustering with multiresolution peak detection. In 2005 IEEE International Conference on Electro Information Technology [1626978] (2005 IEEE International Conference on Electro Information Technology; Vol. 2005).