Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble

Pan Su, Tianhua Chen, Weifeng Xu, Xuqiang Shao, Hongtao Wang, Yitian Zhao

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

Abstract

Conventional clustering approaches partition a set of objects into a certain (some can automatically detect the number of clusters such as DBScan) number of clusters. During the partitioning process, the clusters of objects are produced where each object is assigned to one cluster. On the other hand, the dominantset-based clustering provides a formalisation of clusters by sequentially searching for individual clusters in the set of objects. The resultant clusters do not necessarily form a partition of the set. With the popularity of clustering ensemble, graph-based consensus approaches have been proposed with promising results achieved, many of which are based on the partition of the graph. In this paper, a dominant-set-based consensus method for fuzzy-c-means-based clustering ensemble is proposed. Different from traditional graph-based consensus techniques, the graph generated by the fuzzy clusters are grouped on the basis of the extracted dominant sets. The proposed approach employs a similarity relation to denote the links between component clusters from which the final clusters of ensemble are derived with the extracted dominant set. The proposed method is tested on benchmark data sets against several alternative ensemble methods for fuzzy c-means. The results of experiment show that the proposed dominant-set-based clustering ensemble method generally achieves higher accuracy than its competitors.
Original languageEnglish
Title of host publicationProceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018
PublisherIEEE
Pages85-90
Number of pages6
Volume1
ISBN (Electronic)9781538652145
ISBN (Print)9781538652152, 9781538652121
DOIs
Publication statusPublished - 12 Nov 2018
Event2018 International Conference on Machine Learning and Cybernetics - Crowne Plaza Chengdu City Center, Chengdu, China
Duration: 15 Jul 201818 Jul 2018
http://www.icmlc.com/ICMLC/welcome.html (Link to Conference Website)

Conference

Conference2018 International Conference on Machine Learning and Cybernetics
Abbreviated titleICMLC
CountryChina
CityChengdu
Period15/07/1818/07/18
Internet address

Fingerprint

Experiments

Cite this

Su, P., Chen, T., Xu, W., Shao, X., Wang, H., & Zhao, Y. (2018). Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble. In Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018 (Vol. 1, pp. 85-90). [18232967] IEEE. https://doi.org/10.1109/ICMLC.2018.8526927
Su, Pan ; Chen, Tianhua ; Xu, Weifeng ; Shao, Xuqiang ; Wang, Hongtao ; Zhao, Yitian. / Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble. Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018. Vol. 1 IEEE, 2018. pp. 85-90
@inproceedings{4fd6aa04e9354bfc87a8ba1d9bfa190d,
title = "Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble",
abstract = "Conventional clustering approaches partition a set of objects into a certain (some can automatically detect the number of clusters such as DBScan) number of clusters. During the partitioning process, the clusters of objects are produced where each object is assigned to one cluster. On the other hand, the dominantset-based clustering provides a formalisation of clusters by sequentially searching for individual clusters in the set of objects. The resultant clusters do not necessarily form a partition of the set. With the popularity of clustering ensemble, graph-based consensus approaches have been proposed with promising results achieved, many of which are based on the partition of the graph. In this paper, a dominant-set-based consensus method for fuzzy-c-means-based clustering ensemble is proposed. Different from traditional graph-based consensus techniques, the graph generated by the fuzzy clusters are grouped on the basis of the extracted dominant sets. The proposed approach employs a similarity relation to denote the links between component clusters from which the final clusters of ensemble are derived with the extracted dominant set. The proposed method is tested on benchmark data sets against several alternative ensemble methods for fuzzy c-means. The results of experiment show that the proposed dominant-set-based clustering ensemble method generally achieves higher accuracy than its competitors.",
keywords = "Clustering, Clustering ensemble, Consensus function, Dominant set, Fuzzy c-means",
author = "Pan Su and Tianhua Chen and Weifeng Xu and Xuqiang Shao and Hongtao Wang and Yitian Zhao",
year = "2018",
month = "11",
day = "12",
doi = "10.1109/ICMLC.2018.8526927",
language = "English",
isbn = "9781538652152",
volume = "1",
pages = "85--90",
booktitle = "Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018",
publisher = "IEEE",

}

Su, P, Chen, T, Xu, W, Shao, X, Wang, H & Zhao, Y 2018, Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble. in Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018. vol. 1, 18232967, IEEE, pp. 85-90, 2018 International Conference on Machine Learning and Cybernetics, Chengdu, China, 15/07/18. https://doi.org/10.1109/ICMLC.2018.8526927

Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble. / Su, Pan ; Chen, Tianhua; Xu, Weifeng; Shao, Xuqiang ; Wang, Hongtao; Zhao, Yitian.

Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018. Vol. 1 IEEE, 2018. p. 85-90 18232967.

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

TY - GEN

T1 - Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble

AU - Su, Pan

AU - Chen, Tianhua

AU - Xu, Weifeng

AU - Shao, Xuqiang

AU - Wang, Hongtao

AU - Zhao, Yitian

PY - 2018/11/12

Y1 - 2018/11/12

N2 - Conventional clustering approaches partition a set of objects into a certain (some can automatically detect the number of clusters such as DBScan) number of clusters. During the partitioning process, the clusters of objects are produced where each object is assigned to one cluster. On the other hand, the dominantset-based clustering provides a formalisation of clusters by sequentially searching for individual clusters in the set of objects. The resultant clusters do not necessarily form a partition of the set. With the popularity of clustering ensemble, graph-based consensus approaches have been proposed with promising results achieved, many of which are based on the partition of the graph. In this paper, a dominant-set-based consensus method for fuzzy-c-means-based clustering ensemble is proposed. Different from traditional graph-based consensus techniques, the graph generated by the fuzzy clusters are grouped on the basis of the extracted dominant sets. The proposed approach employs a similarity relation to denote the links between component clusters from which the final clusters of ensemble are derived with the extracted dominant set. The proposed method is tested on benchmark data sets against several alternative ensemble methods for fuzzy c-means. The results of experiment show that the proposed dominant-set-based clustering ensemble method generally achieves higher accuracy than its competitors.

AB - Conventional clustering approaches partition a set of objects into a certain (some can automatically detect the number of clusters such as DBScan) number of clusters. During the partitioning process, the clusters of objects are produced where each object is assigned to one cluster. On the other hand, the dominantset-based clustering provides a formalisation of clusters by sequentially searching for individual clusters in the set of objects. The resultant clusters do not necessarily form a partition of the set. With the popularity of clustering ensemble, graph-based consensus approaches have been proposed with promising results achieved, many of which are based on the partition of the graph. In this paper, a dominant-set-based consensus method for fuzzy-c-means-based clustering ensemble is proposed. Different from traditional graph-based consensus techniques, the graph generated by the fuzzy clusters are grouped on the basis of the extracted dominant sets. The proposed approach employs a similarity relation to denote the links between component clusters from which the final clusters of ensemble are derived with the extracted dominant set. The proposed method is tested on benchmark data sets against several alternative ensemble methods for fuzzy c-means. The results of experiment show that the proposed dominant-set-based clustering ensemble method generally achieves higher accuracy than its competitors.

KW - Clustering

KW - Clustering ensemble

KW - Consensus function

KW - Dominant set

KW - Fuzzy c-means

UR - http://www.scopus.com/inward/record.url?scp=85058009429&partnerID=8YFLogxK

U2 - 10.1109/ICMLC.2018.8526927

DO - 10.1109/ICMLC.2018.8526927

M3 - Conference contribution

SN - 9781538652152

SN - 9781538652121

VL - 1

SP - 85

EP - 90

BT - Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018

PB - IEEE

ER -

Su P, Chen T, Xu W, Shao X, Wang H, Zhao Y. Dominant-Set-Based Consensus for Fuzzy C-Means Clustering Ensemble. In Proceedings of 2018 International Conference on Machine Learning and Cybernetics, ICMLC 2018. Vol. 1. IEEE. 2018. p. 85-90. 18232967 https://doi.org/10.1109/ICMLC.2018.8526927