Skip to main content

Table 1 Performance assessment of the proposed GSOM-based PUL approach, Baseline and OCSVM using Similarity Feature Representation1 (SFR1) and Similarity Feature Representation2 (SFR2)

From: Positive-Unlabeled Learning for inferring drug interactions based on heterogeneous attributes

    Baseline OCSVM GSOM-based PUL
Cross validation SFR1 Precision 0.628 0.584 0.951
   Recall 0.448 0.499 0.861
   F1-score 0.523 0.537 0.904
  SFR2 Precision 0.823 0.622 0.974
   Recall 0.850 0.436 0.975
   F1-score 0.836 0.511 0.974