Home > Research > Publications & Outputs > A Distance-Type-Insensitive Clustering Approach

Electronic data

  • ASOC1

    Rights statement: This is the author’s version of a work that was acceptedfor publication in Applied Soft Computing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Applied Soft Computing, 77, 2019 DOI: 10.1016/j.asoc.2019.01.028

    Accepted author manuscript, 2 MB, PDF document

    Embargo ends: 31/01/20

    Available under license: CC BY-NC-ND: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

Links

Text available via DOI:

View graph of relations

A Distance-Type-Insensitive Clustering Approach

Research output: Contribution to journalJournal article

Published
<mark>Journal publication date</mark>1/04/2019
<mark>Journal</mark>Applied Soft Computing
Volume77
Number of pages13
Pages (from-to)622-634
Publication statusPublished
Early online date31/01/19
Original languageEnglish

Abstract

In this paper, we offer a method aiming to minimise the role of distance metric used in clustering. It is well known that the types of distance metric used in clustering algorithms heavily influence the end results, and also makes the algorithms sensitive to imbalanced attribute scales. To solve these problems, a new clustering algorithm using the per-attribute ranking operating mechanism is proposed in this paper. Ranking is a rarely used discrete, nonlinear operator by other clustering algorithms. However, it also has unique advantages over the dominantly used continuous operators. The proposed algorithm is based on the rankings of the data samples in terms of their spatial separation and is able to provide a more objective clustering result compared with the alternatives. Numerical examples on benchmark datasets prove the validity and effectiveness of the proposed concept and principles.

Bibliographic note

This is the author’s version of a work that was acceptedfor publication in Applied Soft Computing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Applied Soft Computing, 77, 2019 DOI: 10.1016/j.asoc.2019.01.028