Skip to main content

Table 1 Median of WAFs with alternative methodologies and with evoKGsim for the different PPI datasets

From: Evolving knowledge graph similarity for supervised learning in complex biomedical domains

DatasetSSMSingle and static combinationsExhaustive searchDecisionevoKGsim
(#interactions) BPCCMFAvgMaxCombinationsTrees 
STRING-ECSimGIC0.6480.8220.6700.8250.8140.8250.8040.826
(2245)ResnikMax0.6700.8190.6410.8060.8260.8170.8840.864
 ResnikBMA0.6610.8280.6420.8310.8480.8320.8370.849
STRING-DMSimGIC0.8910.8800.7910.8900.8910.9270.8550.936
(550)ResnikMax0.9100.8990.7990.9270.9270.9360.9170.937
 ResnikBMA0.9280.8710.7940.9360.9180.9630.9270.945
BIND-SCSimGIC0.8490.8310.7150.8540.8400.8680.8300.876
(1366)ResnikMax0.8830.8450.7750.9040.9080.9230.8900.923
 ResnikBMA0.8640.8420.7540.9010.8680.9080.8720.901
DIP/MIPS-SCSimGIC0.8110.7760.6900.8030.7790.8180.7540.825
(13807)ResnikMax0.8450.7980.7030.8350.8380.8540.8400.849
 ResnikBMA0.8200.7880.6980.8350.8220.8420.7800.843
STRING-SCSimGIC0.8020.7640.6840.8040.7800.8140.7660.817
(30384)ResnikMax0.8250.7880.6820.8340.8260.8390.8430.843
 ResnikBMA0.8180.7840.6780.8370.8170.8370.7930.838
DIP-HSSimGIC0.8400.7460.6980.8230.7680.8570.7990.861
(2739)ResnikMax0.8920.8290.7700.8850.8670.9140.8940.894
 ResnikBMA0.8740.7730.7540.8760.8110.8720.8670.881
STRING-HSSimGIC0.8240.7690.7000.8130.7860.8230.7740.830
(6912)ResnikMax0.8480.7630.7230.8500.8110.8680.8500.867
 ResnikBMA0.8510.7920.7250.8610.8150.8700.8160.876
GRID/HPRD-unbal-HSSimGIC0.6860.6520.6210.6850.6640.7010.6210.694
(31320)ResnikMax0.7180.6740.6550.7290.7020.7340.7030.734
 ResnikBMA0.7170.6780.6460.7370.6970.7420.6620.742
GRID/HPRD-bal-HSSimGIC0.6470.6300.6180.6720.6470.6740.5900.673
(31349)ResnikMax0.6560.6020.5900.6480.6360.6640.6360.654
 ResnikBMA0.6520.6400.5970.6730.6590.6740.6040.677
  1. In bold, the best result for each dataset-SSM pair. The median WAF achieved for each baseline is underlined when evoKGsim significantly outperforms the baseline (using α=0.01)