A semi-supervised community detection method based on factor graph model

被引：0
|
作者：
Huang L.-W. [1 ]
Li C.-P. [1 ]
Zhang H.-S. [2 ]
Liu Y.-C. [3 ]
Li D.-Y. [4 ]
Liu Y.-B. [1 ]
机构：
[1] Beijing Institute of Remote Sensing, Beijing
[2] Institute of National Defense Information, Wuhan
[3] Chinese Institute of Command and Control, Beijing
[4] Institute of Electronic System Engineering, Beijing
来源：
Huang, Li-Wei (dr_huanglw@163.com) | 1600年 / Science Press卷 / 42期
基金：
中国国家自然科学基金;
关键词：
Factor graph; Probability reasoning; Semi-supervised community detection; Social networks; Social networks analysis;
D O I：
10.16383/j.aas.2016.c150261
中图分类号：
学科分类号：
摘要：
Community detection is an important research direction of social network analysis. Most of the current studies focused on automated community detection. However, in networks having missing data or noise, the ability for an automated community detection algorithm to discover true community structures may degrade rapidly with the increase of noise. On the other hand, semi-supervised community detection provides a feasible way for solving the above problem by incorporating priori information into the community detection process. In this paper, based on the factor graph model, by incorporating the priori information into a unified probabilistic framework, we propose a factor graph-based semi-supervised community detection method. We evaluate the method with three different genres of real datasets (Zachary, Dolphins and DBLP). Experiments indicate that incorporating priori information into the community detection process can improve the prediction accuracy significantly. Compared with a latest semi-supervised community detection algorithm (semi-supervised spin-glass model), the F-measure of our method is on average improved by 6.34%, 16.36% and 12.13% in the three datasets. Copyright © 2016 Acta Automatica Sinica. All rights reserved.
引用
收藏
页码：1520 / 1531
页数：11
相关论文
共 33 条
  • [1] Fortunato S., Community detection in graphs, Physics Reports, 486, 3-5, pp. 75-174, (2010)
  • [2] Eaton E., Mansbach R., A spin-glass model for semi-supervised community detection, Proceedings of the 26th AAAI Conference on Artificial Intelligence, pp. 900-906, (2012)
  • [3] Newman M.E.J., Girvan M., Finding and evaluating community structure in networks, Physical Review E, 69, 2, (2004)
  • [4] Good B.H., De Montjoye Y.A., Clauset A., Performance of modularity maximization in practical contexts, Physical Review E, 81, 4, (2010)
  • [5] Yang Z., Tang J., Li J.Z., Yang W.J., Social community analysis via a factor graph model, IEEE Intelligent Systems, 26, 3, pp. 58-65, (2011)
  • [6] Girvan M., Newman M.E.J., Community structure in social and biological networks, Proceedings of the National Academy of Sciences of the United States of America, 99, 12, pp. 7821-7826, (2002)
  • [7] Brandes U., Delling D., Gaertler M., Gorke R., Hoefer M., Nikoloski Z., Wagner D., On modularity clustering, IEEE Transactions on Knowledge and Data Engineering, 20, 2, pp. 172-188, (2008)
  • [8] Newman M.E.J., Fast algorithm for detecting community structure in networks, Physical Review E, 69, 6, (2004)
  • [9] Duch J., Arenas A., Community detection in complex networks using extremal optimization, Physical Review E, 72, 2, (2005)
  • [10] Guimera R., Amaral L.A.N., Functional cartography of complex metabolic networks, Nature, 433, 7028, pp. 895-900, (2005)
1234