Skip to main content

Table 2 Training and test sets and number of protein pairs respectively used in each experiment

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

Training setNo. of pairsTest setNo. of pairs
S6912D+Gub+Gb47307
D2739S+Gub+Gb51480
Gb31349D+S+Gub22870
Gub31320D+S+Gb22899
S+Gb+Gub69581D2115
D+Gb+Gub65408S5037
S+D9651Gb+Gub44929
Gb+Gub62669S+D7239
S+Gb38261D+Gub17746
D+Gub34059S+Gb20697
S+Gub38232D+Gb17771
D+Gb34088S+Gub20668
  1. The names of the datasets STRING-HS, DIP-HS, GRID/HPRD-unbal-HS, and GRID/HPRD-bal-HS are abbreviated to “S”, “D”, “Gub” and “Gb”, respectively