Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices

Research output: ResearchChapter

Abstract

Semi-supervised clustering uses partially labeled data, as often occurs in practical clustering, to obtain a better clustering result. One approach uses hard constraints which specify data that must and cannot be within the same cluster. In this chapter, we propose another approach to semi-supervised clustering with soft pairwise constraints. The clustering method used is fuzzy c-means (FCM), a commonly used fuzzy clustering method. Two previously proposed variants, entropy- regularized relational/kernel fuzzy c-means clustering and indefinite kernel fuzzy c-means clustering algorithm are modified to use the soft constraints. In addition, a method is discussed that propagates pairwise constraints when the given constraints are not sufficient for obtaining the desired clustering result. Using some numerical examples, it is shown that the proposed algorithms obtain better clustering results.

LanguageEnglish
Title of host publicationStudies in Computational Intelligence
PublisherSpringer Verlag
Pages45-61
Number of pages17
Volume671
DOIs
StatePublished - 2017 Jan 1

Publication series

NameStudies in Computational Intelligence
Volume671
ISSN (Print)1860949X

Fingerprint

Fuzzy clustering
Clustering algorithms
Entropy

Keywords

  • Fuzzy c-means
  • Kernel
  • Relational clustering
  • Semi-supervised clustering

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Kanzawa, Y. (2017). Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices. In Studies in Computational Intelligence (Vol. 671, pp. 45-61). (Studies in Computational Intelligence; Vol. 671). Springer Verlag. DOI: 10.1007/978-3-319-47557-8_4

Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices. / Kanzawa, Yuchi.

Studies in Computational Intelligence. Vol. 671 Springer Verlag, 2017. p. 45-61 (Studies in Computational Intelligence; Vol. 671).

Research output: ResearchChapter

Kanzawa, Y 2017, Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices. in Studies in Computational Intelligence. vol. 671, Studies in Computational Intelligence, vol. 671, Springer Verlag, pp. 45-61. DOI: 10.1007/978-3-319-47557-8_4
Kanzawa Y. Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices. In Studies in Computational Intelligence. Vol. 671. Springer Verlag. 2017. p. 45-61. (Studies in Computational Intelligence). Available from, DOI: 10.1007/978-3-319-47557-8_4
Kanzawa, Yuchi. / Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices. Studies in Computational Intelligence. Vol. 671 Springer Verlag, 2017. pp. 45-61 (Studies in Computational Intelligence).
@inbook{0049e28e94a54748b6348d3f0ac4e4f6,
title = "Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices",
abstract = "Semi-supervised clustering uses partially labeled data, as often occurs in practical clustering, to obtain a better clustering result. One approach uses hard constraints which specify data that must and cannot be within the same cluster. In this chapter, we propose another approach to semi-supervised clustering with soft pairwise constraints. The clustering method used is fuzzy c-means (FCM), a commonly used fuzzy clustering method. Two previously proposed variants, entropy- regularized relational/kernel fuzzy c-means clustering and indefinite kernel fuzzy c-means clustering algorithm are modified to use the soft constraints. In addition, a method is discussed that propagates pairwise constraints when the given constraints are not sufficient for obtaining the desired clustering result. Using some numerical examples, it is shown that the proposed algorithms obtain better clustering results.",
keywords = "Fuzzy c-means, Kernel, Relational clustering, Semi-supervised clustering",
author = "Yuchi Kanzawa",
year = "2017",
month = "1",
doi = "10.1007/978-3-319-47557-8_4",
volume = "671",
series = "Studies in Computational Intelligence",
publisher = "Springer Verlag",
pages = "45--61",
booktitle = "Studies in Computational Intelligence",

}

TY - CHAP

T1 - Semi-supervised fuzzy c-means algorithms by revising dissimilarity/kernel matrices

AU - Kanzawa,Yuchi

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Semi-supervised clustering uses partially labeled data, as often occurs in practical clustering, to obtain a better clustering result. One approach uses hard constraints which specify data that must and cannot be within the same cluster. In this chapter, we propose another approach to semi-supervised clustering with soft pairwise constraints. The clustering method used is fuzzy c-means (FCM), a commonly used fuzzy clustering method. Two previously proposed variants, entropy- regularized relational/kernel fuzzy c-means clustering and indefinite kernel fuzzy c-means clustering algorithm are modified to use the soft constraints. In addition, a method is discussed that propagates pairwise constraints when the given constraints are not sufficient for obtaining the desired clustering result. Using some numerical examples, it is shown that the proposed algorithms obtain better clustering results.

AB - Semi-supervised clustering uses partially labeled data, as often occurs in practical clustering, to obtain a better clustering result. One approach uses hard constraints which specify data that must and cannot be within the same cluster. In this chapter, we propose another approach to semi-supervised clustering with soft pairwise constraints. The clustering method used is fuzzy c-means (FCM), a commonly used fuzzy clustering method. Two previously proposed variants, entropy- regularized relational/kernel fuzzy c-means clustering and indefinite kernel fuzzy c-means clustering algorithm are modified to use the soft constraints. In addition, a method is discussed that propagates pairwise constraints when the given constraints are not sufficient for obtaining the desired clustering result. Using some numerical examples, it is shown that the proposed algorithms obtain better clustering results.

KW - Fuzzy c-means

KW - Kernel

KW - Relational clustering

KW - Semi-supervised clustering

UR - http://www.scopus.com/inward/record.url?scp=85009964393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009964393&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-47557-8_4

DO - 10.1007/978-3-319-47557-8_4

M3 - Chapter

VL - 671

T3 - Studies in Computational Intelligence

SP - 45

EP - 61

BT - Studies in Computational Intelligence

PB - Springer Verlag

ER -