The ROC AUC has an intuitive interpretation: the probability that the score of a randomly sampled 1-labeled item will be higher than a randomly sampled 0-labeled item.
Is there a similar interpretation to the Precision Recall AUC?
Data Science Stack Exchange is a question and answer site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field. It only takes a minute to sign up.Sign up to join this community
No, to my knowledge there is no similar property for the Precision Recall AUC. In fact I think it's not very common to use the PR AUC for evaluation. As far as I know, a PR curve is used mostly to visualize the relation between precision and recall. But there could be usages that I'm not aware of.