HIGHLIGHTS
- who: Davide Boldini from the Center for Functional Protein Assemblies, Technical University of Munich (TUM), Ernstu2011Ottou2011Fischeru2011Strau00dfe, Garching, Germany have published the research: Tuning gradient boosting for imbalanced bioassay modelling with custom loss functions, in the Journal: (JOURNAL)
- what: This approach has the theoretical advantage of weighting the minority class not only according to the class imbalance, but also according to the intrinsic difficulty of the classification problem, which might yield better weights compared to simple class counting statistics . Another advantage is that this approach is function-agnostic, in the sense that it can be implemented . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.