
Postdoc at Technical University of Denmark
pROC 1.7.2 was published this morning. It is a bugfix release that primarily solves various issues with coords and ci.coords. It also warns when computing confidence intervals / roc tests of a ROC curves with AUC == 1 (the CI will always be 11 / ...
pROC 1.7.2 xavier.robin.name
pROC 1.7.2 was published this morning. It is a bugfix release that primarily solves various issues with coords and ci.coords. It also warns when computing confidence intervals / roc tests of a ROC curves with AUC == 1 (the CI will always be 11 /...

Future development: multiclass ROC curves
Postdoc at Technical University of Denmark
This is a question which occurs more and more often: do pROC support multiclass ROC curves?
And the answer is: only a minimal approach is implemented, according to Hand and Till (2001)
method. It gives the mean of all two class AUCs.
Better ... 
Is there any open problem in ROC analysis?
Postgraduate Student at USM
I read some papers said that 'there are still some open problems in ROC analysis', especially refers to multiclass ROC analysis. Can anyone provide some related papers or reports? Thanks!

Power ROC tests, confidence intervals for arbitrary coordinates, speed enhancements: a few of the new features in pROC 1.6
Postdoc at Technical University of Denmark
pROC 1.6 was released, with the following notable new features:
* Power ROC tests
* Confidence intervals for arbitrary coordinates
* Speed enhancements
For more details, see the link below.pROC 1.6 released xavier.robin.name
Two years after the last major release 1.5, pROC 1.6 is finally available. It comes with several major enhancements: This is probably the main feature of this version: power tests for ROC curves. It is now possible to compute sample size, power,...
Jonathan S., Dario V. and 1 other like this

How to use multiclass.roc testing random forest result?
Postgraduate Student at USM
When I use multiclass.roc function, what is the predictor? For instance, I trained a data set by random forest, here is my idea:
rf = randomForest(y~., data)
multiclass.roc(newdata$y, predict(rf, newdata, type = 'prob')$auc )
Is this ...
 1 2
Jonathan S. likes this