Comparing validation results Sample Clauses
Comparing validation results. One additional report is provided to compare the validation results of various prediction algorithms that have been applied to the same dataset (or a range of datasets). This report compares important validation statistics of interest, and automatically performs statistical tests to determine if some algorithms are significantly better than others. We provide an example that compares two state-of-the-art machine learning algorithms: a decision tree algorithm versus a support vector machine. Both prediction algorithms have been evaluated via 5-fold cross- validation on a Blood-Brain-Barrier dataset. The resulting report can be found in the appendix (see 9.3).
