Skip to main content
Cross Validated

Return to Answer

added 151 characters in body
Source Link
Christopher John
Christopher John
  • 258
  • 2
  • 9

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 class distribution sensitive, see https://arxiv.org/ftp/arxiv/papers/1503/1503.06410.pdf. Also consider the ROC and ROC-AUC as an objective way of assessing performance, in which is suited for imbalanced data. In general it is better to look at a curve than rely on a single default cut-off like p=0.5 as the F1 score. I prefer the precision recall gain curve to precision recall, because it is standardised to the baseline. For PR curves it matters a lot which is defined as the positive class and is a deficiency of the approach.

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.

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 class distribution sensitive, see https://arxiv.org/ftp/arxiv/papers/1503/1503.06410.pdf. Also consider the ROC and ROC-AUC as an objective way of assessing performance which is suited for imbalanced data. In general it is better to look at a curve than rely on a single default cut-off like p=0.5 as the F1 score. I prefer the precision recall gain curve to precision recall, because it is standardised to the baseline. For PR curves it matters a lot which is defined as the positive class and is a deficiency of the approach.

deleted 1 character in body
Source Link
Christopher John
Christopher John
  • 258
  • 2
  • 9

I'd suggest avoiding oversampling and undersampling approaches. For a good classifier and performance metric it should not matter and it is introducing biasunscientific. 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 a more unbiasan 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.

I'd suggest avoiding oversampling and undersampling approaches. For a good classifier and performance metric it should not matter and it is introducing bias. 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 a more unbias 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.

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.

Source Link
Christopher John
Christopher John
  • 258
  • 2
  • 9

I'd suggest avoiding oversampling and undersampling approaches. For a good classifier and performance metric it should not matter and it is introducing bias. 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 a more unbias 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.