1
$\begingroup$

Is there any tool that can calculate the AUC value from a ROC curve if I already know how many samples are true positive, true negative, false positive, false negative out of 500 samples?

Specificity and sensitivity are also known to me.

$\endgroup$
  • $\begingroup$ If you don't want to use R, this Python software is quite good too. To install it: pip install CROC. It will provide you with the croc-curve command. Try croc-curve --help to see how to use it. There is a scientific paper about it as well. $\endgroup$ – daruma Sep 21 '18 at 0:50
1
$\begingroup$

Edit: since you apparently do have scores and actual outcomes, you can calculate it. One tool that can do the job would be the pROC package in R. It contains an AUC function that takes as arguments the predicted scores and actual outcomes. Have a look at its documentation http://cran.r-project.org/web/packages/pROC/index.html

-- INITIAL ANSWER --
There is no such tool, because you lack necessary information. You need to have a score for each prediction as well as its true outcome. Without that kind of information, it is impossible to calculate AUC.

$\endgroup$
  • $\begingroup$ Hello sir, i have a score for each prediction as well as its true outcome. $\endgroup$ – user28681 Aug 1 '13 at 7:49
  • $\begingroup$ That changes everything. I've edited my answer to reflect this. $\endgroup$ – mvherweg Aug 1 '13 at 8:01
  • $\begingroup$ Thanx for the reply.It is advantageous to me.. $\endgroup$ – user28681 Aug 1 '13 at 9:26
  • $\begingroup$ Is there any tool present which perform same function as we discussed above. $\endgroup$ – user28681 Aug 1 '13 at 18:26
  • $\begingroup$ @user28681 by computing the confusion matrix (true|false positive|negative counts), you have basically lost the information to compute all possible confusion matrices. You can not possibly compute all the possible confusion matrices from one single confusion matrix. You need the predicted scores for that. $\endgroup$ – Calimo Feb 28 '14 at 14:59

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.