Fig. 5. Distribution of publication counts and number of structurally similar active compounds for top-scoring compounds.
Blue bars represent the number of references associated with the top 20 compounds identified through RF and SVM models. The red bars represent the number of structurally similar active compounds present within the training datasets for these same top-ranked compounds. The compounds were ranked based on their average scores obtained from both models. RF, Random Forest; SVM, Support Vector Machine.
© Korean J Physiol Pharmacol
