An efficient community detection algorithm using greedy surprise maximization

被引:17
作者
Jiang, Yawen [1 ]
Jia, Caiyan [1 ]
Yu, Jian [1 ]
机构
[1] Beijing Jiaotong Univ, Sch Comp & Informat Technol, Beijing Key Lab Traff Data Anal & Min, Beijing 100044, Peoples R China
基金
北京市自然科学基金;
关键词
complex network; community; modularity; resolution limit; surprise; COMPLEX NETWORKS;
D O I
10.1088/1751-8113/47/16/165101
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
Community detection is an important and crucial problem in complex network analysis. Although classical modularity function optimization approaches are widely used for identifying communities, the modularity function (Q) suffers from its resolution limit. Recently, the surprise function (S) was experimentally proved to be better than the Q function. However, up until now, there has been no algorithm available to perform searches to directly determine the maximal surprise values. In this paper, considering the superiority of the S function over the Q function, we propose an efficient community detection algorithm called AGSO (algorithm based on greedy surprise optimization) and its improved version FAGSO (fast-AGSO), which are based on greedy surprise optimization and do not suffer from the resolution limit. In addition, (F) AGSO does not need the number of communities K to be specified in advance. Tests on experimental networks show that (F) AGSO is able to detect optimal partitions in both simple and even more complex networks. Moreover, algorithms based on surprise maximization perform better than those algorithms based on modularity maximization, including Blondel-Guillaume-Lambiotte-Lefebvre (BGLL), Clauset-Newman-Moore (CNM) and the other state-of-the-art algorithms such as Infomap, order statistics local optimization method (OSLOM) and label propagation algorithm (LPA).
引用
收藏
页数:18
相关论文
共 26 条
[11]   Resolution limit in community detection [J].
Fortunato, Santo ;
Barthelemy, Marc .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2007, 104 (01) :36-41
[12]   Community detection in graphs [J].
Fortunato, Santo .
PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS, 2010, 486 (3-5) :75-174
[13]  
Jaccard P., 1901, Bulletin de la Societe Vaudoise des Sciences Naturelles, V37, P241, DOI DOI 10.5169/SEALS-266450
[14]   Finding Statistically Significant Communities in Networks [J].
Lancichinetti, Andrea ;
Radicchi, Filippo ;
Ramasco, Jose J. ;
Fortunato, Santo .
PLOS ONE, 2011, 6 (04)
[15]   Benchmark graphs for testing community detection algorithms [J].
Lancichinetti, Andrea ;
Fortunato, Santo ;
Radicchi, Filippo .
PHYSICAL REVIEW E, 2008, 78 (04)
[16]   Towards real-time community detection in large networks [J].
Leung, Ian X. Y. ;
Hui, Pan ;
Lio, Pietro ;
Crowcroft, Jon .
PHYSICAL REVIEW E, 2009, 79 (06)
[17]   Link prediction in complex networks: A survey [J].
Lue, Linyuan ;
Zhou, Tao .
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS, 2011, 390 (06) :1150-1170
[18]   The bottlenose dolphin community of Doubtful Sound features a large proportion of long-lasting associations - Can geographic isolation explain this unique trait? [J].
Lusseau, D ;
Schneider, K ;
Boisseau, OJ ;
Haase, P ;
Slooten, E ;
Dawson, SM .
BEHAVIORAL ECOLOGY AND SOCIOBIOLOGY, 2003, 54 (04) :396-405
[19]   Tunneling spectroscopy of the superconducting state of URu2Si2 [J].
Maldonado, A. ;
Guillamon, I. ;
Rodrigo, J. G. ;
Suderow, H. ;
Vieira, S. ;
Aoki, D. ;
Flouquet, J. .
PHYSICAL REVIEW B, 2012, 85 (21)
[20]   Defining and identifying communities in networks [J].
Radicchi, F ;
Castellano, C ;
Cecconi, F ;
Loreto, V ;
Parisi, D .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2004, 101 (09) :2658-2663