Hierarchy decomposition pipeline is a supervised machine learning tool that constructs random forest ensembles from data sets with hierarchical class.

Sampling-based class imbalance solutions for multiple-instance classification