riyaeliza123
commented
11 months ago - Due to computational restrictions, we only used 10% of our dataset.
- Even our best model does not give best performance
- class imbalance
- too many false positives, need a measure to deal with that
- we haven't done feature engineering or feature selection
Owl64901