Fig. 5
From: Inverse similarity and reliable negative samples for drug side-effect prediction
Scatter plots of F1-scores for different classifiers using the comprehensive similarity on balanced and imbalanced training sets. The x-axis denotes F1-scores of results based on imbalanced training sets, and the y-axis for balanced training sets. The line “ y=x” on which F1-scores are equal, is the reference line to better visualize the results. Panel (a), (b), (c) and (d) show the scatter plots of F1-scores on balanced and imbalanced training sets using KNN, ELM, SVM and RBF respectively