Data & About

Data settings


Test fbroc with example data

At startup the shiny app will use example dataset “roc.examples” included in fbroc. It includes four different predictors, two continuous and two discrete.

Use your own data

  1. Click on the checkbox “Upload data” to use your own data.
  2. Your data must be stored in tab-delimited text file (Excel can save tables in this format).
  3. The text file must include a column with numerical values. Higher values are assumed to be associated with the positive class.
  4. It must also include a column with the true class labels. The positive class must be TRUE and the negative class FALSE.
  5. Select the correct columns from the dropdown selection above.


If you have problems with or suggestions for either this shiny application or package fbroc please contact me at


This is a Shiny interface for my R-package fbroc, which uses a very fast algorithm implemented in C++ to enable near real-time bootstrapping of the ROC curve and derived performance metrics such as the AUC. This shiny app uses fbroc version 0.3.1.

Current features

  • Very fast bootstrapping of ROC curves.
  • Visualization of confidence regions for the ROC curve.
  • Analysis of the AUC including confidence intervals.
  • Investigate the TPR at a fixed FPR and vice versa.
  • Compare two paired classifiers (not yet supported by the Shiny app)

Planned feature

  • Support for more performance metrics (partial AUC).
  • Help with finding a cutoff optimized for a specific application.

ROC Curve

ROC Curve