ROC curve

A graph of true positive rate versus false positive rate for different classification thresholds in binary classification. The shape of an ROC curve suggests a binary classification model’s ability to separate positive classes from negative classes.¹

Classification threshold가 1.0에서 0.0으로(좌에서 우로) 바뀜에 따라 TPR과 FPR이 어떻게 변하는지 보여주는 시각화 방식.

Etymology

Receiver operating characteristic, is a holdover from World War II radar detection.²

See also

• AUC: A numerical metric that summarizes the ROC curve into a single floating-point value.

Footnotes

1. developers.google.com/machine-learning/glossary#ROC ↩

2. ML crash course - Classification ↩