Fig. 9From: Significant improvement of miRNA target prediction accuracy in large datasets using meta-strategy based on comprehensive voting and artificial neural networksInformation gain compared to the distribution of positive and negative samples in the D4 dataset for (a) miRanda, (b) miRDB, (c) PITA, and (d) TargetScan. X-axis shows the scaled prediction score, y-axis on the left shows the value of information gain, and y-axis on the right shows the distribution of positive samples (red dashed) and negative samples (cyan solid)Back to article page