0
$\begingroup$

What does this roc value mean? How do I interpret it? Are there values which help in inferring it like in case of kappa?enter image description here

$\endgroup$
  • $\begingroup$ Welcome to CV. Whomever downvoted (not me) likely doesn't like your lack of human words or your copy-paste of code. There are answered questions of the general form "explain ROC to grandma" but I think you are asking how it would be interpreted in this case. Tell the story: what motivated you, what problem are you solving, and how did you engage the problem? What do you think? Where did you look for answers? These sorts of things are likely to improve the quality of the question, and get more folks to engage in answering it. $\endgroup$ – EngrStudent - Reinstate Monica Jun 18 at 13:54
0
$\begingroup$

This "ROC area", which you may also hear referred to as "area under the curve" or simply "AUC" is a measure of good your model is. You can read about ROC curves to find out more about it, but I'll tell you the basics:

You'd talk about it in the context of your model making decisions based on a threshold, e.g. when you've got a model that considers a sample "positive" or "negative" based on whether a probability or any other score output by your model is more or less than a certain threshold.

If you were to make a plot of the sensivity vs. the specificity for all values of the threshold, you get a curve. And the area under the curve is this "ROC area" that you're talking about. It'll be a value between 0 and 1. Generally, the higher the area (i.e. the closer to 1), the better.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.