Fuzzy clustering methods for categorical multivariate data based on q-divergence

Tadafumi Kondo, Yuchi Kanzawa

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This paper presents two fuzzy clustering algorithms for categorical multivariate data based on qdivergence. First, this study shows that a conventional method for vectorial data can be explained as regularizing another conventional method using qdivergence. Second, based on the known results that Kullback-Leibler (KL)-divergence is generalized into the q-divergence, and two conventional fuzzy clustering methods for categorical multivariate data adopt KL-divergence, two fuzzy clustering algorithms for categorical multivariate data that are based on qdivergence are derived from two optimization problems built by extending the KL-divergence in these conventional methods to the q-divergence. Through numerical experiments using real datasets, the proposed methods outperform the conventional methods in term of clustering accuracy.

Original languageEnglish
Pages (from-to)524-536
Number of pages13
JournalJournal of Advanced Computational Intelligence and Intelligent Informatics
Volume22
Issue number4
DOIs
Publication statusPublished - 2018 Jul 1

Fingerprint

Fuzzy clustering
Clustering algorithms
Experiments

Keywords

  • Categorical multivariate data
  • Fuzzy clustering
  • KL-divergence
  • Q-divergence

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Fuzzy clustering methods for categorical multivariate data based on q-divergence. / Kondo, Tadafumi; Kanzawa, Yuchi.

In: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 22, No. 4, 01.07.2018, p. 524-536.

Research output: Contribution to journalArticle

@article{20c22560e157427f87974086db462fa7,
title = "Fuzzy clustering methods for categorical multivariate data based on q-divergence",
abstract = "This paper presents two fuzzy clustering algorithms for categorical multivariate data based on qdivergence. First, this study shows that a conventional method for vectorial data can be explained as regularizing another conventional method using qdivergence. Second, based on the known results that Kullback-Leibler (KL)-divergence is generalized into the q-divergence, and two conventional fuzzy clustering methods for categorical multivariate data adopt KL-divergence, two fuzzy clustering algorithms for categorical multivariate data that are based on qdivergence are derived from two optimization problems built by extending the KL-divergence in these conventional methods to the q-divergence. Through numerical experiments using real datasets, the proposed methods outperform the conventional methods in term of clustering accuracy.",
keywords = "Categorical multivariate data, Fuzzy clustering, KL-divergence, Q-divergence",
author = "Tadafumi Kondo and Yuchi Kanzawa",
year = "2018",
month = "7",
day = "1",
doi = "10.20965/jaciii.2018.p0524",
language = "English",
volume = "22",
pages = "524--536",
journal = "Journal of Advanced Computational Intelligence and Intelligent Informatics",
issn = "1343-0130",
publisher = "Fuji Technology Press",
number = "4",

}

TY - JOUR

T1 - Fuzzy clustering methods for categorical multivariate data based on q-divergence

AU - Kondo, Tadafumi

AU - Kanzawa, Yuchi

PY - 2018/7/1

Y1 - 2018/7/1

N2 - This paper presents two fuzzy clustering algorithms for categorical multivariate data based on qdivergence. First, this study shows that a conventional method for vectorial data can be explained as regularizing another conventional method using qdivergence. Second, based on the known results that Kullback-Leibler (KL)-divergence is generalized into the q-divergence, and two conventional fuzzy clustering methods for categorical multivariate data adopt KL-divergence, two fuzzy clustering algorithms for categorical multivariate data that are based on qdivergence are derived from two optimization problems built by extending the KL-divergence in these conventional methods to the q-divergence. Through numerical experiments using real datasets, the proposed methods outperform the conventional methods in term of clustering accuracy.

AB - This paper presents two fuzzy clustering algorithms for categorical multivariate data based on qdivergence. First, this study shows that a conventional method for vectorial data can be explained as regularizing another conventional method using qdivergence. Second, based on the known results that Kullback-Leibler (KL)-divergence is generalized into the q-divergence, and two conventional fuzzy clustering methods for categorical multivariate data adopt KL-divergence, two fuzzy clustering algorithms for categorical multivariate data that are based on qdivergence are derived from two optimization problems built by extending the KL-divergence in these conventional methods to the q-divergence. Through numerical experiments using real datasets, the proposed methods outperform the conventional methods in term of clustering accuracy.

KW - Categorical multivariate data

KW - Fuzzy clustering

KW - KL-divergence

KW - Q-divergence

UR - http://www.scopus.com/inward/record.url?scp=85052023742&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052023742&partnerID=8YFLogxK

U2 - 10.20965/jaciii.2018.p0524

DO - 10.20965/jaciii.2018.p0524

M3 - Article

AN - SCOPUS:85052023742

VL - 22

SP - 524

EP - 536

JO - Journal of Advanced Computational Intelligence and Intelligent Informatics

JF - Journal of Advanced Computational Intelligence and Intelligent Informatics

SN - 1343-0130

IS - 4

ER -