ROC(Receiver operating characteristic) and EER (Equal Error Rate)

ROC

A receiver operating characteristic curve, or ROC curve, is a graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied. From [1]

From [1]

To draw a ROC curve, only the true positive rate (TPR) and false positive rate (FPR) are needed (as functions of some classifier parameter). The TPR defines how many correct positive results occur among all positive samples available during the test. FPR, on the other hand, defines how many incorrect positive results occur among all negative samples available during the test.

From [3]

From [3]

EER

Once you understand ROC curves then the equal error rate loses its mystery. It’s just one way of trying to optimize the trade-off between false positives and false negatives[2]

From [2]

Reference

[1]https://en.wikipedia.org/wiki/Receiver_operating_characteristic

[2]https://www.quora.com/How-can-I-understand-the-EER-Equal-Error-Rate-and-why-we-use-it

[3]https://en.wikipedia.org/wiki/Sensitivity_and_specificity