# Name of error measure based on top |positives| results

I am evaluating a retrieval algorithm. I remember that I read about an error measure which basically entails the quotient of the number of true positives in the top $N_p$ results, where $N_p$ is the number of positives in the set.

$$M = \frac{|TP_{N_p}|}{N_p}$$

I can't remember the name of this measure. Can anyone help me?