Portal de Periódicos da CAPES

PAV and the ROC convex hull

2007; Springer Science+Business Media; Volume: 68; Issue: 1 Linguagem: Inglês

10.1007/s10994-007-5011-0

1573-0565

Tom Fawcett, Alexandru Niculescu-Mizil,

Machine Learning and Data Classification

Classifier calibration is the process of converting classifier scores into reliable probability estimates. Recently, a calibration technique based on isotonic regression has gained attention within machine learning as a flexible and effective way to calibrate classifiers. We show that, surprisingly, isotonic regression based calibration using the Pool Adjacent Violators algorithm is equivalent to the ROC convex hull method.