search

topepo / caret

caret (Classification And Regression Training) R package that contains misc functions for training and plotting classification and regression models
http://topepo.github.io/caret/index.html
1.61k stars 632 forks source link

confusionMatrix: Precision == Positive Predictive Value and Recall == Sensitivity #1115

Open msberends opened 4 years ago

msberends commented 4 years ago

Hi, thanks for the great work on this package!

I noticed that the confusionMatrix() returns the Precision and the Positive Predictive Value, these are actually the same (and should be). Same goes for the Recall and the Sensitivity (True Positive Rate), those are identical as well. So I guess those values could be removed from the calculation.

On the other hand, I would like some additional derivations that are mentioned here: https://en.wikipedia.org/wiki/Confusion_matrix, like the False Negative Rate (FNR), the False Positive Rate (FPR), the False Discovery Rate (FDR) and the False Omission Rate (FOR). Could that be easily added? I could create a PR if you like.

Hu-statistics commented 1 year ago

Are the values of the sensitivity and the specificity in the outcome of confusionMatrix() reversed?