• Acta Physica Sinica
  • Vol. 69, Issue 8, 088901-1 (2020)
Suo-Yi Tan1, Ming-Ze Qi2, Jun Wu3、*, and Xin Lu1、*
Author Affiliations
  • 1College of Systems Engineering, National University of Defense Technology, Changsha 410073, China
  • 2College of Liberal Arts and Sciences, National University of Defense Technology, Changsha 410073, China
  • 3International Academic Center of Complex Systems, Beijing Normal Univerdity, Zhuhai 519087, China
  • show less
    DOI: 10.7498/aps.69.20191817 Cite this Article
    Suo-Yi Tan, Ming-Ze Qi, Jun Wu, Xin Lu. Link predictability of complex network from spectrum perspective[J]. Acta Physica Sinica, 2020, 69(8): 088901-1 Copy Citation Text show less
    Illustration of link prediction problem
    Fig. 1. Illustration of link prediction problem
    The histograms of eigenvalues of random sacle-free networks
    Fig. 2. The histograms of eigenvalues of random sacle-free networks
    The link predictability of model network versus with various
    Fig. 3. The link predictability of model network versus with various
    The link predictability of BA scale-free network and random graph
    Fig. 4. The link predictability of BA scale-free network and random graph
    The performance of prediction algorithms in real networks
    Fig. 5. The performance of prediction algorithms in real networks
    网络$ p $CNAARALPIACARPAKatzRWRACTLRWLHN-II
    BA网络0.9750.4230.3240.2720.4150.2710.2830.5940.4120.1360.5070.0850.003
    随机网络0.5430.0150.0080.0090.0080.00900.0300.0080.0020.02000.001
    Table 1.

    Performance of link prediction algorithms in model networks.

    链路预测算法在模型网络中的表现

    网络VEr$\langle k\rangle$$\langle l\rangle$C
    C_elegans2972148–0.16314.472.460.308
    Windsurfers43336–0.14715.631.700.564
    Adolescent health2539129690.25110.224.520.142
    Jazz19827420.02027.692.210.520
    USAirport157428236–0.11335.873.140.384
    Metabolic4534596–0.22620.292.640.124
    Yeast2375116930.4549.855.100.388
    US powergrid494165940.0032.6720.090.103
    Physicians2411098–0.0569.113.020.552
    Air Traffic Control12262615–0.0154.276.100.063
    Contiguous USA491070.2334.374.260.406
    Email113354510.0789.623.650.166
    King James Bible17739131–0.04810.303.380.163
    Protein Stelzl17066207–0.1917.285.090.006
    Router50226258–0.1382.496.450.033
    Table 2.

    Basic statistics of real networks.

    不同领域真实网络拓扑属性

    网络pCNAARALPIACARPAKatzRWRACTLRWSRW
    C_elegans0.9990.1000.1070.1050.1010.1080.0940.0580.1010.1050.0550.1100.108
    Windsurfers0.9990.3790.3960.4130.3700.3930.3810.2140.3690.3600.2470.4020.426
    Adolescent health0.4220.1030.1030.0880.0890.1010.0940.0030.0880.0530.0080.0420.047
    Jazz1.0000.5020.5230.5420.4890.5350.5170.1330.4890.3520.1680.3420.393
    USAirport0.9980.3330.3360.3640.3320.3320.3300.2800.3320.0870.2940.0760.080
    Metabolic0.9990.1370.1950.2690.1410.1680.1320.1040.1410.1960.0920.2140.215
    Yeast0.9980.1540.1770.2670.1580.1610.1480.0940.1740.0730.2110.0450.059
    US powergrid0.3620.0540.0320.0280.0580.0470.0370.0000.0570.0160.0340.0150.018
    Physicians0.3680.1190.1260.1210.1170.1220.1060.0140.1170.1190.0150.1320.127
    Air Traffic Control0.4800.0360.0240.0180.0370.0210.0250.0070.0370.0020.0150.0020.002
    Contiguous USA0.5400.0960.1300.1320.0050.0000.0000.0120.0040.0670.0530.1330.121
    Email0.9500.1440.1580.1430.1420.1590.1450.0180.1410.0650.0240.0520.051
    King James Bible0.9600.1670.2700.4460.1630.2560.1760.0780.1630.1860.0690.1970.224
    Protein Stelzl0.4410.0010.0020.0010.0010.0020.0060.0140.0010.0060.0130.0060.006
    Router0.5110.0510.0290.0200.0560.0310.0550.0220.0550.0060.1640.0050.005
    Table 3.

    The precision of link prediction algorithms in real networks

    链路预测算法在真实网络中的precision值

    Suo-Yi Tan, Ming-Ze Qi, Jun Wu, Xin Lu. Link predictability of complex network from spectrum perspective[J]. Acta Physica Sinica, 2020, 69(8): 088901-1
    Download Citation