463 0

Dependence clustering, a method revealing community structure with group dependence

Title
Dependence clustering, a method revealing community structure with group dependence
Author
이기천
Keywords
Group dependence; Clustering; Markovian; Community structure; Mutual information
Issue Date
2014-04
Publisher
Elsevier Science B.V
Citation
Knowledge-Based Systems, 2014, 60, P.58-72
Abstract
We propose a clustering method maximizing a new measure called "group dependence." Group dependence quantifies how precise a certain division of a graph is in terms of dependence distance. Built upon statistical dependence measure between points driven by Markovian transitions, group dependence incorporates the geometric structure of input data. Besides capturing degrees of positive dependence and coherence for a group division, group dependence inherently supplies the proposed clustering method with a definite decision on the depth of division. We provide an optimality aspect of the method as theoretical justification in consideration of posterior transition probabilities of input data. Illustrating its procedure using data from a known structure, we demonstrate its performance in the clustering task of real-world data sets, Amazon, DBLP, and YouTube, in comparison with selected clustering algorithms. We show that the proposed method outperforms the selected methods in reasonable settings: in particular, the proposed method surpasses modularity clustering in terms of normalized mutual information. We also show that the proposed method reveals additional insights on community structure detection according to its connectivity scale parameter. (C) 2014 Elsevier B.V. All rights reserved.
URI
http://www.sciencedirect.com/science/article/pii/S095070511400015Xhttp://hdl.handle.net/20.500.11754/48005
ISSN
0950-7051; 1872-7409
DOI
10.1016/j.knosys.2014.01.004
Appears in Collections:
COLLEGE OF ENGINEERING[S](공과대학) > INDUSTRIAL ENGINEERING(산업공학과) > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE