0
$\begingroup$

How to calculate AUC using some formula? What are the parameters required and what formula to use.

For accuracy, $$ \frac{TP+TN}{Total} $$ is this right way to calculate AUC?

| improve this question | |
$\endgroup$
  • $\begingroup$ These are Confusion Matrix metrics, you can see the formula for every metric in wikipedia $\endgroup$ – Pedro Henrique Monforte Apr 16 '19 at 20:55
1
$\begingroup$

For accuracy, TP+TN/total is it right way to calculate?

If your problem is binary classification, then yes.

How to calculate AUC using some formula? What are the parameters required and what formula to use?

Check this question.

| improve this answer | |
$\endgroup$
  • $\begingroup$ What can be the difference in range of value of AUC and accuracy? (0 to 1) Which one is more? and using tp,tn,fp,fn can we calculate AUC without drawing roc curve? $\endgroup$ – Shivanya Apr 16 '19 at 16:20
  • 1
    $\begingroup$ @Shivanya Those would be better as new questions than as comments, but AUC goes from [0.5, 1], with larger values being "better". You do not need to draw an ROC curve to calculate AUC, though it is useful for comparing different decision thresholds. $\endgroup$ – Upper_Case Apr 17 '19 at 18:38

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.