Fig. 7From: Considerations for feature selection using gene pairs and applications in large-scale dataset integration, novel oncogene discovery, and interpretable cancer screeningSimplified tree ensemble learners for ovarian and pancreatic cancer. Circulating miRNA were used to build STELs for (a) ovarian and (b) pancreatic cancer screening. Frequency refers to the frequency of cases in the dataset that satisfy the corresponding rule. Error refers to the error rate of the rule when the corresponding outcome is chosenBack to article page