Explain how a ROC curve works?

SAIKAT HALDER

tech blogger | Posted on 3-Dec-2018

To visualize the performance of a binary classifier, the receiver operating characteristic curve popularly known as ROC curve is used. Making ROC curve is not that simple as one need to be familiar enough with the concepts of true positive and negative and false positive and negative where these concepts are used to compare the results of a test.

It is basically a graph and the curve on x-axis and y-axis shows the relationship between clinical sensitivity and specificity. Even though you cannot see the cut-off each point on ROC represents a chosen cut-off. And when one chooses this cut-off, true positive friction and the false positive friction can be seen.

In medical world ROC curve is made by ranking the values and linking each one of them with the diagnosis. Not only in medical world but ROC curve works positively in many other areas to measure the performance of different multi-class classifiers. The area under ROC curve is highlighted to get the results of the test. Though there are many other ways of measuring the performance or diagnosis it is proven as a best way where accurate result or answer can be expected.

ref links: https://www.google.co.in/url?sa=i&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwj_rP7M8fzeAhVDfn0KHXzzANQQjhx6BAgBEAM&url=http%3A%2F%2Fwww.anaesthetist.com%2Fmnm%2Fstats%2Froc%2F&psig=AOvVaw26qZC8GbXCvK0AutoRjgSG&ust=1543693664387424

video courtesy: www.youtube.com