Is there a 'precision' or 'recall' metric for True Negatives?

Question

Noob question..

Accuracy is defined as percentage of predictions that are correct.

Accuracy = (True Positives + True Negatives)/(All classifications)

Precision is defined as percentage of positive predictions that are correct.

Precision = (True Positives)/(True Positives + False Positives)

Is there a term for the percentage of negative predictions that are correct:

UnPrecision = (True Negatives)/(True Negatives+ False Negatives) ?

Similarly, how about:

UnRecall = (True Negatives)/(True Negatives + False Positives) ?

If not, why not?

Or is it just a case of semantics, where we just swap the terms positive and negative if needed, so we can use the same forumla?

Answer 1

You seem to be looking for the Negative Predictive Value and the True Negative Rate.

Answer 2

Your "UnRecall" is usually called "specificity" and my suggested name for "UnPrecision" is "exclusion" but I don't know if that is a standard. The other piece is "sensitivity" which is TP/(TP+FN).