Are there any general guidelines for dealing with imbalanced data through upsampling/downsampling?

This Google developer guide suggests performing downsampling with upweighting, but for the most part I've found upsampling usually works better in practice (some corroboration).

Is there any clear consensus or empirical study of what works in practice, or when to use which? Does it matter which classification algorithm you use?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.