Abstract

In traditional co-clustering, the only basis for the clustering task is a given relationship matrix, describing the strengths of the relationships between pairs of elements in the different domains. Relying on this single input matrix, co-clustering discovers relationships holding among groups of elements from the two input domains. In many real life applications, on the other hand, other background knowledge or metadata about one or more of the two input domain dimensions may be available and, if leveraged properly, such metadata might play a significant role in the effectiveness of the co-clustering process. How additional metadata affects co-clustering, however, depends on how the process is modified to be context-aware. In this paper, we propose, compare, and evaluate three alternative strategies (metadata-driven, metadata-constrained, and metadata-injected co-clustering) for embedding available contextual knowledge into the co-clustering process. Experimental results show that it is possible to leverage the available metadata in discovering contextually-relevant co-clusters, without significant overheads in terms of information theoretical co-cluster quality or execution cost.

Original languageEnglish (US)
Pages (from-to)209-239
Number of pages31
JournalJournal of Intelligent Information Systems
Volume38
Issue number1
DOIs
StatePublished - Feb 2012

Fingerprint

Metadata
Costs

Keywords

  • Co-clustering
  • Concept alignment
  • Constraints
  • Context-aware clustering
  • Metadata

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems
  • Hardware and Architecture
  • Computer Networks and Communications
  • Software

Cite this

On context-aware co-clustering with metadata support. / Schifanella, Claudio; Sapino, Maria Luisa; Candan, Kasim.

In: Journal of Intelligent Information Systems, Vol. 38, No. 1, 02.2012, p. 209-239.

Research output: Contribution to journalArticle

Schifanella, Claudio ; Sapino, Maria Luisa ; Candan, Kasim. / On context-aware co-clustering with metadata support. In: Journal of Intelligent Information Systems. 2012 ; Vol. 38, No. 1. pp. 209-239.
@article{4fdb1377d14542d487cc2808dd1aebdd,
title = "On context-aware co-clustering with metadata support",
abstract = "In traditional co-clustering, the only basis for the clustering task is a given relationship matrix, describing the strengths of the relationships between pairs of elements in the different domains. Relying on this single input matrix, co-clustering discovers relationships holding among groups of elements from the two input domains. In many real life applications, on the other hand, other background knowledge or metadata about one or more of the two input domain dimensions may be available and, if leveraged properly, such metadata might play a significant role in the effectiveness of the co-clustering process. How additional metadata affects co-clustering, however, depends on how the process is modified to be context-aware. In this paper, we propose, compare, and evaluate three alternative strategies (metadata-driven, metadata-constrained, and metadata-injected co-clustering) for embedding available contextual knowledge into the co-clustering process. Experimental results show that it is possible to leverage the available metadata in discovering contextually-relevant co-clusters, without significant overheads in terms of information theoretical co-cluster quality or execution cost.",
keywords = "Co-clustering, Concept alignment, Constraints, Context-aware clustering, Metadata",
author = "Claudio Schifanella and Sapino, {Maria Luisa} and Kasim Candan",
year = "2012",
month = "2",
doi = "10.1007/s10844-011-0151-x",
language = "English (US)",
volume = "38",
pages = "209--239",
journal = "Journal of Intelligent Information Systems",
issn = "0925-9902",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - On context-aware co-clustering with metadata support

AU - Schifanella, Claudio

AU - Sapino, Maria Luisa

AU - Candan, Kasim

PY - 2012/2

Y1 - 2012/2

N2 - In traditional co-clustering, the only basis for the clustering task is a given relationship matrix, describing the strengths of the relationships between pairs of elements in the different domains. Relying on this single input matrix, co-clustering discovers relationships holding among groups of elements from the two input domains. In many real life applications, on the other hand, other background knowledge or metadata about one or more of the two input domain dimensions may be available and, if leveraged properly, such metadata might play a significant role in the effectiveness of the co-clustering process. How additional metadata affects co-clustering, however, depends on how the process is modified to be context-aware. In this paper, we propose, compare, and evaluate three alternative strategies (metadata-driven, metadata-constrained, and metadata-injected co-clustering) for embedding available contextual knowledge into the co-clustering process. Experimental results show that it is possible to leverage the available metadata in discovering contextually-relevant co-clusters, without significant overheads in terms of information theoretical co-cluster quality or execution cost.

AB - In traditional co-clustering, the only basis for the clustering task is a given relationship matrix, describing the strengths of the relationships between pairs of elements in the different domains. Relying on this single input matrix, co-clustering discovers relationships holding among groups of elements from the two input domains. In many real life applications, on the other hand, other background knowledge or metadata about one or more of the two input domain dimensions may be available and, if leveraged properly, such metadata might play a significant role in the effectiveness of the co-clustering process. How additional metadata affects co-clustering, however, depends on how the process is modified to be context-aware. In this paper, we propose, compare, and evaluate three alternative strategies (metadata-driven, metadata-constrained, and metadata-injected co-clustering) for embedding available contextual knowledge into the co-clustering process. Experimental results show that it is possible to leverage the available metadata in discovering contextually-relevant co-clusters, without significant overheads in terms of information theoretical co-cluster quality or execution cost.

KW - Co-clustering

KW - Concept alignment

KW - Constraints

KW - Context-aware clustering

KW - Metadata

UR - http://www.scopus.com/inward/record.url?scp=84856212922&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856212922&partnerID=8YFLogxK

U2 - 10.1007/s10844-011-0151-x

DO - 10.1007/s10844-011-0151-x

M3 - Article

VL - 38

SP - 209

EP - 239

JO - Journal of Intelligent Information Systems

JF - Journal of Intelligent Information Systems

SN - 0925-9902

IS - 1

ER -