Recall Precision Tradeoff

High
Sensitivity means less false negative

High
Precision means less false positive

Recall Precision Curve

Max Precision observed at Recall R or higher.

The precision-recall curve is dataset dependent, particularly on positive labels, because the true positive rate at threshold 0 cannot be exactly 0 but is bounded by the ratio of positive labels to the total number of samples in the dataset.

Mean Average Precision (MAP)

More weight to the top items

Only binary relevance

Most standard among the TREC community is MAP which provides a single-figure measure of quality across recall levels.

Precision and recall

A precision-recall curve plots precision as a function of recall; usually precision will decrease as the recall increases. Alternatively, values for one measure can be compared for a fixed level at the other measure (e.g. precision at a recall level of 0.75) or both are combined into a single measure

https://en.wikipedia.org/wiki/Precision_and_recall#:~:text=A%20precision%2Drecall%20curve%20plots,combined%20into%20a%20single%20measure

Recall Precision Tradeoff

Recall Precision Curve

Mean Average Precision (MAP)

Recommendations