An Efficient Selection of Initial Cluster in K-Means Using Entropy and Co-efficient of Variation for High Dimensional Data
Main Article Content
Abstract
K-means clustering is a method of cluster analysis which aims to partition n observations into k-clusters in which each observation belongs to the cluster with the nearest mean. For high dimensional dataset K-means can’t give better cluster output, feature selection methods are required to remove irrelevant features. But the proposed algorithm selects primary and secondary axes based on means and variations of individual column. Here entropy value is combined with mean and variation of axes selection. This may find the axes which are more relevant and highly ranked. The integration of these two techniques achieves feature selection and initial centroid selection simultaneously. Real-time high dimensional datasets are used for experiments to show that the proposed algorithm provides better results for high-dimensional dataset.
Keywords: K-means algorithm, initial cluster centers, high dimensional dataset, error percentage, entropy.
Downloads
Article Details
COPYRIGHT
Submission of a manuscript implies: that the work described has not been published before, that it is not under consideration for publication elsewhere; that if and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the publisher.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
- The journal allows the author(s) to retain publishing rights without restrictions.
- The journal allows the author(s) to hold the copyright without restrictions.