Decoupling Decision-Making in Fraud Prevention through Classifier Calibration for Business Logic Action

Open paper

Benchmark	Model	Rank	Results
fraud-detection-on-baf-base	CatBoost	–	Recall @ 5% FPR: 52.4%
fraud-detection-on-baf-base	LightGBM	–	Recall @ 5% FPR: 54.3%
fraud-detection-on-baf-base	MLP–NN	–	Recall @ 5% FPR: 49.6%