Skip to main content

Table 4 Comparison of results for binary relevance (baseline) and classifiers chain, by ten-fold cross-validation

From: Selectivity profiling of BCRP versus P-gp inhibition: from automated collection of polypharmacology data to multi-label learning

Algorithms Macro-accuracy Macro-MCC Macro-AUC
Binary relevance, Logistic Regression 0.812 0.594 0.793
Classifiers chain, Logisitic Regression 0.812 0.594 0.793
Binary relevance, RandomForest 0.835 0.641 0.808
Classifiers chain, RandomForest 0.836 0.643 0.809
Binary relevance, SVM 0.766 0.504 0.749
Classifiers chain, SVM 0.767 0.504 0.750
  1. In italic letters, the model that gives the best results