A class-weighted Global AdaBoost implementation for rare-event fraud detection over 6.36 million PaySim transactions. The project uses PySpark for feature preparation, weighted decision-tree training ...
Boosting algorithms consistently beat the classical baselines, but the margin over Random Forest shrinks on small/noisy data (Wine Quality) and widens on larger, mixed-feature data (Adult Income).
Some results have been hidden because they may be inaccessible to you
Show inaccessible results