Acta Physica Sinica, Volume. 69, Issue 16, 168901-1(2020)

Link prediction model based on dynamic network representation

Zhong-Ming Han1...2,*, Sheng-Nan Li1, Chen-Ye Zheng1, Da-Gao Duan1 and Wei-Jie Yang1 |Show fewer author(s)
Author Affiliations
  • 1College of Computer and Information Engineering, Beijing Technology and Business University, Beijing 100048, China
  • 2Beijing Key Laboratory of Food Safety Big Data Technology, Beijing Technology and Business University, Beijing 100048, China
  • show less
    Figures & Tables(13)
    Schematic diagram of dynamic network.
    The architecture of link prediction model based on dynamic network representation.
    Time interval based LSTM unit.
    Schematic diagram of node neighborhood sampling.
    Node neighborhood update unit.
    comparison diagram on each data set. (a) UCI dataset; (b) DNC dataset; (b) Wikipedia dataset; (d) Enron dataset
    comparison diagram on each data set. (a) UCI dataset; (b) DNC dataset; (b) Wikipedia dataset; (d) Enron dataset
    Recall@k comparison diagram of the variants of DNRLP. (a) UCI dataset; (b) DNC dataset; (c) Wikipedia dataset; (d) Enron dataset.
    MRR results of different training rates. (a) DNC dataset; (b) Enron dataset
    • Table 1.

      Information diffusion algorithm.

      信息扩散算法

      View table
      View in Article

      Table 1.

      Information diffusion algorithm.

      信息扩散算法

      输入: 新增链接 $ {e}_{ij}\in {E}_{{\rm{new}}} $, 随机游走长度 $ L $
      输出: 随机游走序列 $ R $
      1) For $ {e}_{ij} $ in $ {E}_{{\rm{new}}} $ do:
      2) For $ v $ in $ {e}_{ij} $ do:
      3)   $ m=\mathrm{ }0 $
      4)  While $ m < L $ do
      5)   初始化权重分布 $ P $
      6)   For $ u $ in $ {N}_{v} $ do
      7)    根据(13)式计算 $ {f}_{\rm{s}}\left({u}_{\rm{u}}, {u}_{v}\right) $, 加入 $ P $
      8)   End for
      9)   根据 $ P $选择下一个节点 $u^\prime$加入 ${R}_{v}$
      10)    $ m=m+1 $
      11)    $ v=u' $
      12)  End while
      13)  将 $ {R}_{v} $加入 $ R $
      14) End for
      15) End for
    • Table 2.

      Dynamic network data details.

      动态网络数据详细信息

      View table
      View in Article

      Table 2.

      Dynamic network data details.

      动态网络数据详细信息

      数据集节点数边数时间/d聚类系数/%
      UCI1899598351945.68
      DNC2029392649828.90
      Wikipedia1219241228454647630.000837
      Enron384413175146311404.96
    • Table 3.

      Experimental environment setup information.

      实验环境设置信息

      View table
      View in Article

      Table 3.

      Experimental environment setup information.

      实验环境设置信息

      项目设置数量
      操作系统Ubuntu 16.041
      CPUIntel®i7-5280K, 6 核, 12线程1
      硬盘512GB PLEXTOR®PX-512M6Pro SSD1
      内存Kingston®8GB DDR4 24008
      重要程序包Python 3.71
      深度学习 框架 PyTorch1
    • Table 4.

      Link prediction MRR results comparison.

      链接预测MRR结果对比

      View table
      View in Article

      Table 4.

      Link prediction MRR results comparison.

      链接预测MRR结果对比

      方法UCIDNCWikipediaEnron
      Logistic Regression0.005 10.020 90.003 70.005 2
      SVM0.003 20.018 20.002 10.002 9
      Node2Vec0.004 70.019 70.003 50.003 9
      GCN0.015 90.048 40.010 10.017 6
      GraphSAGE0.016 30.049 70.012 00.018 3
      DynGEM0.015 70.028 40.010 80.014 7
      GCN-GAN0.020 10.050 40.014 90.021 5
      DDNE0.014 20.026 80.009 60.011 6
      DNRLP0.035 10.053 90.018 70.036 3
    Tools

    Get Citation

    Copy Citation Text

    Zhong-Ming Han, Sheng-Nan Li, Chen-Ye Zheng, Da-Gao Duan, Wei-Jie Yang. Link prediction model based on dynamic network representation[J]. Acta Physica Sinica, 2020, 69(16): 168901-1

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Jul. 29, 2019

    Accepted: --

    Published Online: Jan. 4, 2021

    The Author Email:

    DOI:10.7498/aps.69.20191162

    Topics