Linear semi-supervised projection clustering by transferred centroid regularization

Bin Tong, Hao Shao, Bin Hui Chou, Einoshin Suzuki

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We propose a novel method, called Semi-supervised Projection Clustering in Transfer Learning (SPCTL), where multiple source domains and one target domain are assumed. Traditional semi-supervised projection clustering methods hold the assumption that the data and pairwise constraints are all drawn from the same domain. However, many related data sets with different distributions are available in real applications. The traditional methods thus can not be directly extended to such a scenario. One major challenging issue is how to exploit constraint knowledge from multiple source domains and transfer it to the target domain where all the data are unlabeled. To handle this difficulty, we are motivated to construct a common subspace where the difference in distributions among domains can be reduced. We also invent a transferred centroid regularization, which acts as a bridge to transfer the constraint knowledge to the target domain, to formulate this geometric structure formed by the centroids from different domains. Extensive experiments on both synthetic and benchmark data sets show the effectiveness of our method.

Original languageEnglish
Pages (from-to)461-490
Number of pages30
JournalJournal of Intelligent Information Systems
Volume39
Issue number2
DOIs
Publication statusPublished - Oct 1 2012

Fingerprint

Experiments

All Science Journal Classification (ASJC) codes

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

Cite this

Linear semi-supervised projection clustering by transferred centroid regularization. / Tong, Bin; Shao, Hao; Chou, Bin Hui; Suzuki, Einoshin.

In: Journal of Intelligent Information Systems, Vol. 39, No. 2, 01.10.2012, p. 461-490.

Research output: Contribution to journalArticle

@article{467be49a854048259f6a30f8bb5b161c,
title = "Linear semi-supervised projection clustering by transferred centroid regularization",
abstract = "We propose a novel method, called Semi-supervised Projection Clustering in Transfer Learning (SPCTL), where multiple source domains and one target domain are assumed. Traditional semi-supervised projection clustering methods hold the assumption that the data and pairwise constraints are all drawn from the same domain. However, many related data sets with different distributions are available in real applications. The traditional methods thus can not be directly extended to such a scenario. One major challenging issue is how to exploit constraint knowledge from multiple source domains and transfer it to the target domain where all the data are unlabeled. To handle this difficulty, we are motivated to construct a common subspace where the difference in distributions among domains can be reduced. We also invent a transferred centroid regularization, which acts as a bridge to transfer the constraint knowledge to the target domain, to formulate this geometric structure formed by the centroids from different domains. Extensive experiments on both synthetic and benchmark data sets show the effectiveness of our method.",
author = "Bin Tong and Hao Shao and Chou, {Bin Hui} and Einoshin Suzuki",
year = "2012",
month = "10",
day = "1",
doi = "10.1007/s10844-012-0198-3",
language = "English",
volume = "39",
pages = "461--490",
journal = "Journal of Intelligent Information Systems",
issn = "0925-9902",
publisher = "Springer Netherlands",
number = "2",

}

TY - JOUR

T1 - Linear semi-supervised projection clustering by transferred centroid regularization

AU - Tong, Bin

AU - Shao, Hao

AU - Chou, Bin Hui

AU - Suzuki, Einoshin

PY - 2012/10/1

Y1 - 2012/10/1

N2 - We propose a novel method, called Semi-supervised Projection Clustering in Transfer Learning (SPCTL), where multiple source domains and one target domain are assumed. Traditional semi-supervised projection clustering methods hold the assumption that the data and pairwise constraints are all drawn from the same domain. However, many related data sets with different distributions are available in real applications. The traditional methods thus can not be directly extended to such a scenario. One major challenging issue is how to exploit constraint knowledge from multiple source domains and transfer it to the target domain where all the data are unlabeled. To handle this difficulty, we are motivated to construct a common subspace where the difference in distributions among domains can be reduced. We also invent a transferred centroid regularization, which acts as a bridge to transfer the constraint knowledge to the target domain, to formulate this geometric structure formed by the centroids from different domains. Extensive experiments on both synthetic and benchmark data sets show the effectiveness of our method.

AB - We propose a novel method, called Semi-supervised Projection Clustering in Transfer Learning (SPCTL), where multiple source domains and one target domain are assumed. Traditional semi-supervised projection clustering methods hold the assumption that the data and pairwise constraints are all drawn from the same domain. However, many related data sets with different distributions are available in real applications. The traditional methods thus can not be directly extended to such a scenario. One major challenging issue is how to exploit constraint knowledge from multiple source domains and transfer it to the target domain where all the data are unlabeled. To handle this difficulty, we are motivated to construct a common subspace where the difference in distributions among domains can be reduced. We also invent a transferred centroid regularization, which acts as a bridge to transfer the constraint knowledge to the target domain, to formulate this geometric structure formed by the centroids from different domains. Extensive experiments on both synthetic and benchmark data sets show the effectiveness of our method.

UR - http://www.scopus.com/inward/record.url?scp=84857869442&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857869442&partnerID=8YFLogxK

U2 - 10.1007/s10844-012-0198-3

DO - 10.1007/s10844-012-0198-3

M3 - Article

AN - SCOPUS:84857869442

VL - 39

SP - 461

EP - 490

JO - Journal of Intelligent Information Systems

JF - Journal of Intelligent Information Systems

SN - 0925-9902

IS - 2

ER -