Home > Research > Publications & Outputs > Data density based clustering
View graph of relations

Data density based clustering

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Data density based clustering. / Hyde, Richard; Angelov, Plamen.
Computational Intelligence (UKCI), 2014 14th UK Workshop on. IEEE, 2014. p. 1-7.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Hyde, R & Angelov, P 2014, Data density based clustering. in Computational Intelligence (UKCI), 2014 14th UK Workshop on. IEEE, pp. 1-7, UKCI2014, Bradford, United Kingdom, 8/09/14. https://doi.org/10.1109/UKCI.2014.6930157

APA

Hyde, R., & Angelov, P. (2014). Data density based clustering. In Computational Intelligence (UKCI), 2014 14th UK Workshop on (pp. 1-7). IEEE. https://doi.org/10.1109/UKCI.2014.6930157

Vancouver

Hyde R, Angelov P. Data density based clustering. In Computational Intelligence (UKCI), 2014 14th UK Workshop on. IEEE. 2014. p. 1-7 doi: 10.1109/UKCI.2014.6930157

Author

Hyde, Richard ; Angelov, Plamen. / Data density based clustering. Computational Intelligence (UKCI), 2014 14th UK Workshop on. IEEE, 2014. pp. 1-7

Bibtex

@inproceedings{1f33059483b444f7bf5caf5abd728fc9,
title = "Data density based clustering",
abstract = "A new, data density based approach to clustering is presented which automatically determines the number of clusters. By using RDE for each data sample the number of calculations is significantly reduced in offline mode and, further, the method is suitable for online use. The clusters allow a different diameter per feature/ dimension creating hyper-ellipsoid clusters which are axis-orthogonal. This results in a greater differentiation between clusters where the clusters are highly asymmetrical. We illustrate this with 3 standard data sets, 1 artificial dataset and a large real dataset to demonstrate comparable results to Subtractive, Hierarchical, K-Means, ELM and DBScan clustering techniques. Unlike subtractive clustering we do not iteratively calculate P however. Unlike hierarchical we do not need O(N2) distances to be calculated and a cut-off threshold to be defined. Unlike k-means we do not need to predefine the number of clusters.Using the RDE equations to calculate the densities the algorithm is efficient, and requires no iteration to approach the optimal result. We compare the proposed algorithm to k-means, subtractive, hierarchical, ELM and DBScan clustering with respect to several criteria. The results demonstrate the validity of the proposed approach.",
keywords = "clustering, incremental clustering, big data",
author = "Richard Hyde and Plamen Angelov",
year = "2014",
month = sep,
day = "8",
doi = "10.1109/UKCI.2014.6930157",
language = "English",
isbn = "9781479955381",
pages = "1--7",
booktitle = "Computational Intelligence (UKCI), 2014 14th UK Workshop on",
publisher = "IEEE",
note = "UKCI2014 ; Conference date: 08-09-2014 Through 10-09-2014",

}

RIS

TY - GEN

T1 - Data density based clustering

AU - Hyde, Richard

AU - Angelov, Plamen

PY - 2014/9/8

Y1 - 2014/9/8

N2 - A new, data density based approach to clustering is presented which automatically determines the number of clusters. By using RDE for each data sample the number of calculations is significantly reduced in offline mode and, further, the method is suitable for online use. The clusters allow a different diameter per feature/ dimension creating hyper-ellipsoid clusters which are axis-orthogonal. This results in a greater differentiation between clusters where the clusters are highly asymmetrical. We illustrate this with 3 standard data sets, 1 artificial dataset and a large real dataset to demonstrate comparable results to Subtractive, Hierarchical, K-Means, ELM and DBScan clustering techniques. Unlike subtractive clustering we do not iteratively calculate P however. Unlike hierarchical we do not need O(N2) distances to be calculated and a cut-off threshold to be defined. Unlike k-means we do not need to predefine the number of clusters.Using the RDE equations to calculate the densities the algorithm is efficient, and requires no iteration to approach the optimal result. We compare the proposed algorithm to k-means, subtractive, hierarchical, ELM and DBScan clustering with respect to several criteria. The results demonstrate the validity of the proposed approach.

AB - A new, data density based approach to clustering is presented which automatically determines the number of clusters. By using RDE for each data sample the number of calculations is significantly reduced in offline mode and, further, the method is suitable for online use. The clusters allow a different diameter per feature/ dimension creating hyper-ellipsoid clusters which are axis-orthogonal. This results in a greater differentiation between clusters where the clusters are highly asymmetrical. We illustrate this with 3 standard data sets, 1 artificial dataset and a large real dataset to demonstrate comparable results to Subtractive, Hierarchical, K-Means, ELM and DBScan clustering techniques. Unlike subtractive clustering we do not iteratively calculate P however. Unlike hierarchical we do not need O(N2) distances to be calculated and a cut-off threshold to be defined. Unlike k-means we do not need to predefine the number of clusters.Using the RDE equations to calculate the densities the algorithm is efficient, and requires no iteration to approach the optimal result. We compare the proposed algorithm to k-means, subtractive, hierarchical, ELM and DBScan clustering with respect to several criteria. The results demonstrate the validity of the proposed approach.

KW - clustering

KW - incremental clustering

KW - big data

U2 - 10.1109/UKCI.2014.6930157

DO - 10.1109/UKCI.2014.6930157

M3 - Conference contribution/Paper

SN - 9781479955381

SP - 1

EP - 7

BT - Computational Intelligence (UKCI), 2014 14th UK Workshop on

PB - IEEE

T2 - UKCI2014

Y2 - 8 September 2014 through 10 September 2014

ER -