I don't quite understand how p&r better than accuracy?
limitation: unable to handle skewed data?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
0
Precision and recall contain more information than accuracy. Precision and recall give you information on how many items your model missed, and how much junk it returned. You can create a precision-recall curve based on your model to evaluate different cut points and select one based on the goal of your model.
Also look into sensitivity (the same as recall) and specificity, which will give information on the ability of your model to identify true negatives.
