Return to Revisions

I'd suggest avoiding oversampling and undersampling approaches. For a good classifier and performance metric it should not matter and it is unscientific. The F1 score is bias towards the majority class and is distribution sensitive, see https://arxiv.org/ftp/arxiv/papers/1503/1503.06410.pdf. Also consider the ROC-AUC as an objective way of assessing performance, in general it is better to look at a curve than rely on a single cut-off like p=0.5 as the F1 score.