Difference between revisions of "Precision and recall"

From RecSysWiki
Jump to navigation Jump to search
 
Line 1: Line 1:
'''Precision''' or '''positive predictive value''' is a measure for exactness and '''recall''' or '''sensitivity''' is a measure of completeness.
+
'''Precision''' or '''positive predictive value''' is a [[measure]] for exactness and '''recall''' or '''sensitivity''' is a measure of [[completeness]].
 
They range from 0 to 1 and the best possible value for both of them is 1.  
 
They range from 0 to 1 and the best possible value for both of them is 1.  
  
Line 6: Line 6:
 
:<math>\text{recall} = \frac{tp}{tp + fn},</math>
 
:<math>\text{recall} = \frac{tp}{tp + fn},</math>
  
:where <math>tp</math> is the number of true positives, <math>fp</math> the number of false positives, and <math>fn</math> the number of false negatives.
+
:where <math>tp</math> is the number of [[true positives]], <math>fp</math> the number of false positives, and <math>fn</math> the number of [[false negatives]].
  
 
== External links ==
 
== External links ==

Latest revision as of 10:02, 4 December 2014

Precision or positive predictive value is a measure for exactness and recall or sensitivity is a measure of completeness. They range from 0 to 1 and the best possible value for both of them is 1.

<math>\text{precision} = \frac{tp}{tp + fp}</math>
<math>\text{recall} = \frac{tp}{tp + fn},</math>
where <math>tp</math> is the number of true positives, <math>fp</math> the number of false positives, and <math>fn</math> the number of false negatives.

External links