Encyclopedia of Machine Learning

2010 Edition
| Editors: Claude Sammut, Geoffrey I. Webb

Sensitivity and Specificity

  • Kai Ming Ting
Reference work entry
DOI: https://doi.org/10.1007/978-0-387-30164-8_752

Definition

Sensitivity and specificity are two measures used together in some domains to measure the predictive performance of a classification model or a diagnostic test. For example, to measure the effectiveness of a diagnostic test in the medical domain, sensitivity measures the fraction of people with disease (i.e., positive examples) who have a positive test result; and specificity measures the fraction of people without disease (i.e., negative examples) who have a negative test result. They are defined with reference to a special case of the  confusion matrix, with two classes, one designated the positive class, and the other the negative class, as indicated in Table  1.
Sensitivity and Specificity. Table 1

The outcomes of classification into positive and negative classes

   

Assigned Class

   
   

Positive

Negative

 
 

Positive

True Positive (TP)

False Negative (FN)

 

Actual Class

Negative

False Positive (FP)

True Negative (TN)

 
This is a preview of subscription content, log in to check access.

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Kai Ming Ting

There are no affiliations available