Efficient nonparametric and asymptotic Bayesian model selection methods for attributed graph clustering

Handle URI:
http://hdl.handle.net/10754/622931
Title:
Efficient nonparametric and asymptotic Bayesian model selection methods for attributed graph clustering
Authors:
Xu, Zhiqiang; Cheng, James; Xiao, Xiaokui; Fujimaki, Ryohei; Muraoka, Yusuke
Abstract:
Attributed graph clustering, also known as community detection on attributed graphs, attracts much interests recently due to the ubiquity of attributed graphs in real life. Many existing algorithms have been proposed for this problem, which are either distance based or model based. However, model selection in attributed graph clustering has not been well addressed, that is, most existing algorithms assume the cluster number to be known a priori. In this paper, we propose two efficient approaches for attributed graph clustering with automatic model selection. The first approach is a popular Bayesian nonparametric method, while the second approach is an asymptotic method based on a recently proposed model selection criterion, factorized information criterion. Experimental results on both synthetic and real datasets demonstrate that our approaches for attributed graph clustering with automatic model selection significantly outperform the state-of-the-art algorithm.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Xu Z, Cheng J, Xiao X, Fujimaki R, Muraoka Y (2017) Efficient nonparametric and asymptotic Bayesian model selection methods for attributed graph clustering. Knowledge and Information Systems. Available: http://dx.doi.org/10.1007/s10115-017-1030-8.
Publisher:
Springer Nature
Journal:
Knowledge and Information Systems
Issue Date:
16-Feb-2017
DOI:
10.1007/s10115-017-1030-8
Type:
Article
ISSN:
0219-1377; 0219-3116
Sponsors:
The authors would like to thank the anonymous reviewers of the paper for their valuable comments that help significantly improve the quality of the paper.
Additional Links:
http://link.springer.com/article/10.1007/s10115-017-1030-8
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorXu, Zhiqiangen
dc.contributor.authorCheng, Jamesen
dc.contributor.authorXiao, Xiaokuien
dc.contributor.authorFujimaki, Ryoheien
dc.contributor.authorMuraoka, Yusukeen
dc.date.accessioned2017-02-28T11:32:08Z-
dc.date.available2017-02-28T11:32:08Z-
dc.date.issued2017-02-16en
dc.identifier.citationXu Z, Cheng J, Xiao X, Fujimaki R, Muraoka Y (2017) Efficient nonparametric and asymptotic Bayesian model selection methods for attributed graph clustering. Knowledge and Information Systems. Available: http://dx.doi.org/10.1007/s10115-017-1030-8.en
dc.identifier.issn0219-1377en
dc.identifier.issn0219-3116en
dc.identifier.doi10.1007/s10115-017-1030-8en
dc.identifier.urihttp://hdl.handle.net/10754/622931-
dc.description.abstractAttributed graph clustering, also known as community detection on attributed graphs, attracts much interests recently due to the ubiquity of attributed graphs in real life. Many existing algorithms have been proposed for this problem, which are either distance based or model based. However, model selection in attributed graph clustering has not been well addressed, that is, most existing algorithms assume the cluster number to be known a priori. In this paper, we propose two efficient approaches for attributed graph clustering with automatic model selection. The first approach is a popular Bayesian nonparametric method, while the second approach is an asymptotic method based on a recently proposed model selection criterion, factorized information criterion. Experimental results on both synthetic and real datasets demonstrate that our approaches for attributed graph clustering with automatic model selection significantly outperform the state-of-the-art algorithm.en
dc.description.sponsorshipThe authors would like to thank the anonymous reviewers of the paper for their valuable comments that help significantly improve the quality of the paper.en
dc.publisherSpringer Natureen
dc.relation.urlhttp://link.springer.com/article/10.1007/s10115-017-1030-8en
dc.rightsThe final publication is available at Springer via http://dx.doi.org/10.1007/s10115-017-1030-8en
dc.subjectAttributed graph clusteringen
dc.subjectModel selectionen
dc.subjectDirichlet processen
dc.subjectFactorized information criterionen
dc.titleEfficient nonparametric and asymptotic Bayesian model selection methods for attributed graph clusteringen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalKnowledge and Information Systemsen
dc.eprint.versionPost-printen
dc.contributor.institutionDepartment of Computer Science and Engineering, The Chinese University of Hong Kong, Shatin, Chinaen
dc.contributor.institutionSchool of Computer Engineering, Nanyang Technological University, Singapore, Singaporeen
dc.contributor.institutionNEC Laboratories America, Cupertino, USAen
dc.contributor.institutionNEC Laboratories Japan, Nakahara-ku, Kawasaki-shi, Japanen
kaust.authorXu, Zhiqiangen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.