A unified metric for categorical and numerical attributes in data clustering

Yiu Ming CHEUNG; Hong Jia

doi:10.1007/978-3-642-37456-2_12

A unified metric for categorical and numerical attributes in data clustering

Yiu Ming CHEUNG, Hong Jia

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

11 Citations (Scopus)

Abstract

Most of the existing clustering approaches are applicable to purely numerical or categorical data only, but not both. In general, it is a nontrivial task to perform clustering on mixed data composed of numerical and categorical attributes because there exists an awkward gap between the similarity metrics for categorical and numerical data. This paper therefore presents a general clustering framework based on the concept of object-cluster similarity and gives a unified similarity metric which can be simply applied to the data with categorical, numerical, or mixed attributes. Accordingly, an iterative clustering algorithm is developed, whose efficacy is experimentally demonstrated on different benchmark data sets.

Original language	English
Title of host publication	Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings
Pages	135-146
Number of pages	12
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-37456-2_12
Publication status	Published - 2013
Event	17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013 - Gold Coast, QLD, Australia Duration: 14 Apr 2013 → 17 Apr 2013

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	7819 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013
Country/Territory	Australia
City	Gold Coast, QLD
Period	14/04/13 → 17/04/13

Scopus Subject Areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-642-37456-2_12

Cite this

CHEUNG, Y. M., & Jia, H. (2013). A unified metric for categorical and numerical attributes in data clustering. In Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings (PART 2 ed., pp. 135-146). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7819 LNAI, No. PART 2). https://doi.org/10.1007/978-3-642-37456-2_12

CHEUNG, Yiu Ming ; Jia, Hong. / A unified metric for categorical and numerical attributes in data clustering. Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings. PART 2. ed. 2013. pp. 135-146 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{77d0f9507b8849ebafa1441e671b828e,

title = "A unified metric for categorical and numerical attributes in data clustering",

abstract = "Most of the existing clustering approaches are applicable to purely numerical or categorical data only, but not both. In general, it is a nontrivial task to perform clustering on mixed data composed of numerical and categorical attributes because there exists an awkward gap between the similarity metrics for categorical and numerical data. This paper therefore presents a general clustering framework based on the concept of object-cluster similarity and gives a unified similarity metric which can be simply applied to the data with categorical, numerical, or mixed attributes. Accordingly, an iterative clustering algorithm is developed, whose efficacy is experimentally demonstrated on different benchmark data sets.",

author = "CHEUNG, {Yiu Ming} and Hong Jia",

note = "Copyright: Copyright 2014 Elsevier B.V., All rights reserved.; 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013 ; Conference date: 14-04-2013 Through 17-04-2013",

year = "2013",

doi = "10.1007/978-3-642-37456-2_12",

language = "English",

isbn = "9783642374555",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "135--146",

booktitle = "Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings",

edition = "PART 2",

}

CHEUNG, YM & Jia, H 2013, A unified metric for categorical and numerical attributes in data clustering. in Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 7819 LNAI, pp. 135-146, 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013, Gold Coast, QLD, Australia, 14/04/13. https://doi.org/10.1007/978-3-642-37456-2_12

A unified metric for categorical and numerical attributes in data clustering. / CHEUNG, Yiu Ming; Jia, Hong.
Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings. PART 2. ed. 2013. p. 135-146 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7819 LNAI, No. PART 2).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - A unified metric for categorical and numerical attributes in data clustering

AU - CHEUNG, Yiu Ming

AU - Jia, Hong

PY - 2013

Y1 - 2013

N2 - Most of the existing clustering approaches are applicable to purely numerical or categorical data only, but not both. In general, it is a nontrivial task to perform clustering on mixed data composed of numerical and categorical attributes because there exists an awkward gap between the similarity metrics for categorical and numerical data. This paper therefore presents a general clustering framework based on the concept of object-cluster similarity and gives a unified similarity metric which can be simply applied to the data with categorical, numerical, or mixed attributes. Accordingly, an iterative clustering algorithm is developed, whose efficacy is experimentally demonstrated on different benchmark data sets.

AB - Most of the existing clustering approaches are applicable to purely numerical or categorical data only, but not both. In general, it is a nontrivial task to perform clustering on mixed data composed of numerical and categorical attributes because there exists an awkward gap between the similarity metrics for categorical and numerical data. This paper therefore presents a general clustering framework based on the concept of object-cluster similarity and gives a unified similarity metric which can be simply applied to the data with categorical, numerical, or mixed attributes. Accordingly, an iterative clustering algorithm is developed, whose efficacy is experimentally demonstrated on different benchmark data sets.

UR - http://www.scopus.com/inward/record.url?scp=84893571444&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-37456-2_12

DO - 10.1007/978-3-642-37456-2_12

M3 - Conference proceeding

AN - SCOPUS:84893571444

SN - 9783642374555

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 135

EP - 146

BT - Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings

T2 - 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013

Y2 - 14 April 2013 through 17 April 2013

ER -

CHEUNG YM, Jia H. A unified metric for categorical and numerical attributes in data clustering. In Advances in Knowledge Discovery and Data Mining - 17th Pacific-Asia Conference, PAKDD 2013, Proceedings. PART 2 ed. 2013. p. 135-146. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-37456-2_12

A unified metric for categorical and numerical attributes in data clustering

Abstract

Publication series

Conference

Scopus Subject Areas

Access to Document

Other files and links

Fingerprint

Cite this