Coupled clustering: A method for detecting structural correspondence

Zvika Marx; Ido Dagan; Joachim M. Buhmann; Eli Shamir

Coupled clustering: A method for detecting structural correspondence

Zvika Marx, Ido Dagan, Joachim M. Buhmann, Eli Shamir

Department of Computer Science

Research output: Contribution to journal › Article › peer-review

27 Scopus citations

Abstract

This paper proposes a new paradigm and a computational framework for revealing equivalencies (analogies) between sub-structures of distinct composite systems that are initially represented by unstructured data sets. For this purpose, we introduce and investigate a variant of traditional data clustering, termed coupled clustering, which outputs a configuration of corresponding subsets of two such representative sets. We apply our method to synthetic as well as textual data. Its achievements in detecting topical correspondences between textual corpora are evaluated through comparison to performance of human.

Original language	English
Pages (from-to)	747-780
Number of pages	34
Journal	Journal of Machine Learning Research
Volume	3
Issue number	4-5
State	Published - 15 May 2003

Bibliographical note

Funding Information:
We acknowledge the support of the Hungarian state and the European Union TAMOP-4.2.2A-11/1/KONV-2012-0072 and TAMOP-4.1.1C-12/1/KONV-2012-0017. This paper was also supported by the J?nos Bolyai Research Scholarship of the Hungarian Academy of Sciences.

Keywords

Clustering
Data mining in texts
Natural language processing
Structure mapping
Unsupervised learning

Cite this

@article{1b659dc4ab704004a6018d95a2462646,

title = "Coupled clustering: A method for detecting structural correspondence",

abstract = "This paper proposes a new paradigm and a computational framework for revealing equivalencies (analogies) between sub-structures of distinct composite systems that are initially represented by unstructured data sets. For this purpose, we introduce and investigate a variant of traditional data clustering, termed coupled clustering, which outputs a configuration of corresponding subsets of two such representative sets. We apply our method to synthetic as well as textual data. Its achievements in detecting topical correspondences between textual corpora are evaluated through comparison to performance of human.",

keywords = "Clustering, Data mining in texts, Natural language processing, Structure mapping, Unsupervised learning",

author = "Zvika Marx and Ido Dagan and Buhmann, {Joachim M.} and Eli Shamir",

note = "Funding Information: We acknowledge the support of the Hungarian state and the European Union TAMOP-4.2.2A-11/1/KONV-2012-0072 and TAMOP-4.1.1C-12/1/KONV-2012-0017. This paper was also supported by the J?nos Bolyai Research Scholarship of the Hungarian Academy of Sciences.",

year = "2003",

month = may,

day = "15",

language = "אנגלית",

volume = "3",

pages = "747--780",

journal = "Journal of Machine Learning Research",

issn = "1532-4435",

publisher = "Microtome Publishing",

number = "4-5",

}

TY - JOUR

T1 - Coupled clustering

T2 - A method for detecting structural correspondence

AU - Marx, Zvika

AU - Dagan, Ido

AU - Buhmann, Joachim M.

AU - Shamir, Eli

N1 - Funding Information: We acknowledge the support of the Hungarian state and the European Union TAMOP-4.2.2A-11/1/KONV-2012-0072 and TAMOP-4.1.1C-12/1/KONV-2012-0017. This paper was also supported by the J?nos Bolyai Research Scholarship of the Hungarian Academy of Sciences.

PY - 2003/5/15

Y1 - 2003/5/15

N2 - This paper proposes a new paradigm and a computational framework for revealing equivalencies (analogies) between sub-structures of distinct composite systems that are initially represented by unstructured data sets. For this purpose, we introduce and investigate a variant of traditional data clustering, termed coupled clustering, which outputs a configuration of corresponding subsets of two such representative sets. We apply our method to synthetic as well as textual data. Its achievements in detecting topical correspondences between textual corpora are evaluated through comparison to performance of human.

AB - This paper proposes a new paradigm and a computational framework for revealing equivalencies (analogies) between sub-structures of distinct composite systems that are initially represented by unstructured data sets. For this purpose, we introduce and investigate a variant of traditional data clustering, termed coupled clustering, which outputs a configuration of corresponding subsets of two such representative sets. We apply our method to synthetic as well as textual data. Its achievements in detecting topical correspondences between textual corpora are evaluated through comparison to performance of human.

KW - Clustering

KW - Data mining in texts

KW - Natural language processing

KW - Structure mapping

KW - Unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=0141607829&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0141607829

SN - 1532-4435

VL - 3

SP - 747

EP - 780

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

IS - 4-5

ER -

Coupled clustering: A method for detecting structural correspondence

Abstract

Bibliographical note

Keywords

Other files and links

Fingerprint

Cite this