An Evolutionary Multiobjective Approach for Community Discovery in Dynamic Networks

被引:196
作者
Folino, Francesco [1 ]
Pizzuti, Clara [1 ]
机构
[1] Natl Res Council Italy, CNR, Inst High Performance Comp & Networking, ICAR, I-87036 Arcavacata Di Rende, Italy
关键词
Evolutionary clustering; complex networks; dynamic networks; community discovery; GENETIC ALGORITHM; DIRICHLET PROCESS;
D O I
10.1109/TKDE.2013.131
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The discovery of evolving communities in dynamic networks is an important research topic that poses challenging tasks. Evolutionary clustering is a recent framework for clustering dynamic networks that introduces the concept of temporal smoothness inside the community structure detection method. Evolutionary-based clustering approaches try to maximize cluster accuracy with respect to incoming data of the current time step, and minimize clustering drift from one time step to the successive one. In order to optimize both these two competing objectives, an input parameter that controls the preference degree of a user towards either the snapshot quality or the temporal quality is needed. In this paper the detection of communities with temporal smoothness is formulated as a multiobjective problem and a method based on genetic algorithms is proposed. The main advantage of the algorithm is that it automatically provides a solution representing the best trade-off between the accuracy of the clustering obtained, and the deviation from one time step to the successive. Experiments on synthetic data sets show the very good performance of the method when compared with state-of-the-art approaches.
引用
收藏
页码:1838 / 1852
页数:15
相关论文
共 49 条
  • [1] [Anonymous], 2006, P 12 ACM SIGKDD INT, DOI [10.1145/1150402.1150467, DOI 10.1145/1150402.1150467]
  • [2] [Anonymous], 1989, PROC 3 ANN C GENET A
  • [3] [Anonymous], 2007, P 13 ACM SIGKDD INT, DOI DOI 10.1145/1281192.1281266
  • [4] [Anonymous], 2004, Information Sciences Institute Technical Report
  • [5] [Anonymous], P 13 INT C KNOWL DIS
  • [6] [Anonymous], 2005, MULTICRITERIA OPTIMI
  • [7] [Anonymous], 2006, Proceedings of 12th International Conference on Knowledge Discovery in Data Mining
  • [8] [Anonymous], 2005, P 11 ACM SIGKDD INT
  • [9] An Event-Based Framework for Characterizing the Evolutionary Behavior of Interaction Graphs
    Asur, Sitaram
    Parthasarathy, Srinivasan
    Ucar, Duygu
    [J]. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2009, 3 (04)
  • [10] Berger-Wolf T, 2010, LINK MINING: MODELS, ALGORITHMS, AND APPLICATIONS, P307, DOI 10.1007/978-1-4419-6515-8_12