An Efficient Centroid Selection Algorithm for K-means Clustering

Abstract

This paper, we proposes an algorithm for performing data partitioning along the data axis with the highest variance to improve the accuracy. The data partitioning tries to divide data space into small cells or clusters where inter cluster distance are large and intra cluster distance are small as possible. Cells are partitioned one at a time until the number of cells equals to the predefined number of clusters, K. The centers of the K cells become the initial cluster centers for K-means. The experimental results shows that the proposed algorithm will be more effective and efficient converge to better clustering results than the existing clustering.

Authors and Affiliations

Saranya and Dr. Punithavalli

Keywords

Related Articles

Fractal Image Coding: A New Approach with Block Merging

The most of the fractal coding schemes proposed so far are based on Iterated Function Systems, Range block and domain blocks partition. In this paper, a scheme is proposed which is based on searching the blocks in imag...

SELECTION OF MIXED SAMPLING PLAN WITH CSP- 3(k=i) PLAN AS ATTRIBUTE PLAN INDEXED THROUGH MAPD AND MAAOQ

In this paper a procedure for the construction and selection of the independent mixed sampling plan using MAPD and MAAOQ as quality standards with Continuous Sampling plan of the type CSP-3(k=i) as attribute plan is pr...

HIGHER ORDER MUTATION TESTING (RESULT- EQUIVALENT MUTANTS)

Whenever we make a single change to the original program we get First Order Mutant (FOM). When we apply another single change to FOM we get Second Order Mutant (SOM).On applying another single change to SOM we get Thir...

A STUDY OF NEWSPAPER READERSHIP PATTERNS AMONG URBAN POPULATION OF PUNJAB.

Newspapers have been a source of information to one and all from various years. However, with the proliferation of TV News Channels, Readers' choice of getting news and information has changed. The evolution of new med...

Fuzzy Controller for an Image based Traffic System.

Traffic problems nowadays are increasing because of the growing number of vehicles and the limited resources provided by current infrastructures. The simplest way for controlling a traffic light uses timer for each pha...

Download PDF file
  • EP ID EP26484
  • DOI -
  • Views 351
  • Downloads 8

How To Cite

Saranya and Dr. Punithavalli (2011). An Efficient Centroid Selection Algorithm for K-means Clustering. International Journal of Engineering, Science and Mathematics, 1(3), -. https://europub.co.uk/articles/-A-26484