A density-based noisy graph partitioning algorithm

Jaehong Yu, Seoung Bum Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Clustering analysis can facilitate the extraction of implicit patterns in a dataset and elicit its natural groupings without requiring prior classification information. Numerous researchers have focused recently on graph-based clustering algorithms because their graph structure is useful in modeling the local relationships among observations. These algorithms perform reasonably well in their intended applications. However, no consensus exists about which of them best satisfies all the conditions encountered in a variety of real situations. In this study, we propose a graph-based clustering algorithm based on a novel density-of-graph structure. In the proposed algorithm, a density coefficient defined for each node is used to classify dense and sparse nodes. The main structures of clusters are identified through dense nodes and sparse nodes that are assigned to specific clusters. Experiments on various simulation datasets and benchmark datasets were conducted to examine the properties of the proposed algorithm and to compare its performance with that of existing spectral clustering and modularity-based algorithms. The experimental results demonstrated that the proposed clustering algorithm performed better than its competitors; this was especially true when the cluster structures in the data were inherently noisy and nonlinearly distributed.

Original languageEnglish
JournalNeurocomputing
DOIs
Publication statusAccepted/In press - 2015 Mar 8

Fingerprint

Clustering algorithms
Cluster Analysis
Benchmarking
Research Personnel
Experiments
Datasets

Keywords

  • Clustering algorithm
  • Density coefficient
  • Maximizing connectivity
  • Nonlinearity

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

A density-based noisy graph partitioning algorithm. / Yu, Jaehong; Kim, Seoung Bum.

In: Neurocomputing, 08.03.2015.

Research output: Contribution to journalArticle

@article{9624de271db04bd995263e86d19dede9,
title = "A density-based noisy graph partitioning algorithm",
abstract = "Clustering analysis can facilitate the extraction of implicit patterns in a dataset and elicit its natural groupings without requiring prior classification information. Numerous researchers have focused recently on graph-based clustering algorithms because their graph structure is useful in modeling the local relationships among observations. These algorithms perform reasonably well in their intended applications. However, no consensus exists about which of them best satisfies all the conditions encountered in a variety of real situations. In this study, we propose a graph-based clustering algorithm based on a novel density-of-graph structure. In the proposed algorithm, a density coefficient defined for each node is used to classify dense and sparse nodes. The main structures of clusters are identified through dense nodes and sparse nodes that are assigned to specific clusters. Experiments on various simulation datasets and benchmark datasets were conducted to examine the properties of the proposed algorithm and to compare its performance with that of existing spectral clustering and modularity-based algorithms. The experimental results demonstrated that the proposed clustering algorithm performed better than its competitors; this was especially true when the cluster structures in the data were inherently noisy and nonlinearly distributed.",
keywords = "Clustering algorithm, Density coefficient, Maximizing connectivity, Nonlinearity",
author = "Jaehong Yu and Kim, {Seoung Bum}",
year = "2015",
month = "3",
day = "8",
doi = "10.1016/j.neucom.2015.10.085",
language = "English",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - A density-based noisy graph partitioning algorithm

AU - Yu, Jaehong

AU - Kim, Seoung Bum

PY - 2015/3/8

Y1 - 2015/3/8

N2 - Clustering analysis can facilitate the extraction of implicit patterns in a dataset and elicit its natural groupings without requiring prior classification information. Numerous researchers have focused recently on graph-based clustering algorithms because their graph structure is useful in modeling the local relationships among observations. These algorithms perform reasonably well in their intended applications. However, no consensus exists about which of them best satisfies all the conditions encountered in a variety of real situations. In this study, we propose a graph-based clustering algorithm based on a novel density-of-graph structure. In the proposed algorithm, a density coefficient defined for each node is used to classify dense and sparse nodes. The main structures of clusters are identified through dense nodes and sparse nodes that are assigned to specific clusters. Experiments on various simulation datasets and benchmark datasets were conducted to examine the properties of the proposed algorithm and to compare its performance with that of existing spectral clustering and modularity-based algorithms. The experimental results demonstrated that the proposed clustering algorithm performed better than its competitors; this was especially true when the cluster structures in the data were inherently noisy and nonlinearly distributed.

AB - Clustering analysis can facilitate the extraction of implicit patterns in a dataset and elicit its natural groupings without requiring prior classification information. Numerous researchers have focused recently on graph-based clustering algorithms because their graph structure is useful in modeling the local relationships among observations. These algorithms perform reasonably well in their intended applications. However, no consensus exists about which of them best satisfies all the conditions encountered in a variety of real situations. In this study, we propose a graph-based clustering algorithm based on a novel density-of-graph structure. In the proposed algorithm, a density coefficient defined for each node is used to classify dense and sparse nodes. The main structures of clusters are identified through dense nodes and sparse nodes that are assigned to specific clusters. Experiments on various simulation datasets and benchmark datasets were conducted to examine the properties of the proposed algorithm and to compare its performance with that of existing spectral clustering and modularity-based algorithms. The experimental results demonstrated that the proposed clustering algorithm performed better than its competitors; this was especially true when the cluster structures in the data were inherently noisy and nonlinearly distributed.

KW - Clustering algorithm

KW - Density coefficient

KW - Maximizing connectivity

KW - Nonlinearity

UR - http://www.scopus.com/inward/record.url?scp=84949034773&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949034773&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2015.10.085

DO - 10.1016/j.neucom.2015.10.085

M3 - Article

AN - SCOPUS:84949034773

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -