A model-based approach to attributed graph clustering

Graph clustering, also known as community detection, is a long-standing problem in data mining. However, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage structural and attribute information for clustering attributed graphs becomes a ne...

Full description

Saved in:
Bibliographic Details
Main Authors: Xu, Zhiqiang, Ke, Yiping, Wang, Yi, Cheng, Hong, Cheng, James
Other Authors: School of Computer Engineering
Format: Conference or Workshop Item
Language:English
Published: 2013
Subjects:
Online Access:https://hdl.handle.net/10356/98766
http://hdl.handle.net/10220/12623
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-98766
record_format dspace
spelling sg-ntu-dr.10356-987662020-05-28T07:18:13Z A model-based approach to attributed graph clustering Xu, Zhiqiang Ke, Yiping Wang, Yi Cheng, Hong Cheng, James School of Computer Engineering International Conference on Management of Data (2012) DRNTU::Engineering::Computer science and engineering Graph clustering, also known as community detection, is a long-standing problem in data mining. However, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage structural and attribute information for clustering attributed graphs becomes a new challenge. Most existing works take a distance-based approach. They proposed various distance measures to combine structural and attribute information. In this paper, we consider an alternative view and propose a model-based approach to attributed graph clustering. We develop a Bayesian probabilistic model for attributed graphs. The model provides a principled and natural framework for capturing both structural and attribute aspects of a graph, while avoiding the artificial design of a distance measure. Clustering with the proposed model can be transformed into a probabilistic inference problem, for which we devise an efficient variational algorithm. Experimental results on large real-world datasets demonstrate that our method significantly outperforms the state-of-art distance-based attributed graph clustering method. 2013-07-31T06:34:10Z 2019-12-06T19:59:28Z 2013-07-31T06:34:10Z 2019-12-06T19:59:28Z 2012 2012 Conference Paper Xu, Z., Ke, Y., Wang, Y., Cheng, H., & Cheng, J. (2012). A model-based approach to attributed graph clustering. Proceedings of the 2012 international conference on Management of Data - SIGMOD '12, 505-516. https://hdl.handle.net/10356/98766 http://hdl.handle.net/10220/12623 10.1145/2213836.2213894 en
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic DRNTU::Engineering::Computer science and engineering
spellingShingle DRNTU::Engineering::Computer science and engineering
Xu, Zhiqiang
Ke, Yiping
Wang, Yi
Cheng, Hong
Cheng, James
A model-based approach to attributed graph clustering
description Graph clustering, also known as community detection, is a long-standing problem in data mining. However, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage structural and attribute information for clustering attributed graphs becomes a new challenge. Most existing works take a distance-based approach. They proposed various distance measures to combine structural and attribute information. In this paper, we consider an alternative view and propose a model-based approach to attributed graph clustering. We develop a Bayesian probabilistic model for attributed graphs. The model provides a principled and natural framework for capturing both structural and attribute aspects of a graph, while avoiding the artificial design of a distance measure. Clustering with the proposed model can be transformed into a probabilistic inference problem, for which we devise an efficient variational algorithm. Experimental results on large real-world datasets demonstrate that our method significantly outperforms the state-of-art distance-based attributed graph clustering method.
author2 School of Computer Engineering
author_facet School of Computer Engineering
Xu, Zhiqiang
Ke, Yiping
Wang, Yi
Cheng, Hong
Cheng, James
format Conference or Workshop Item
author Xu, Zhiqiang
Ke, Yiping
Wang, Yi
Cheng, Hong
Cheng, James
author_sort Xu, Zhiqiang
title A model-based approach to attributed graph clustering
title_short A model-based approach to attributed graph clustering
title_full A model-based approach to attributed graph clustering
title_fullStr A model-based approach to attributed graph clustering
title_full_unstemmed A model-based approach to attributed graph clustering
title_sort model-based approach to attributed graph clustering
publishDate 2013
url https://hdl.handle.net/10356/98766
http://hdl.handle.net/10220/12623
_version_ 1681058556760031232