Precision and recall

precisionrecallF-measurePrecision, recallrecall and precisionRecall (information retrieval)recall accuracyrecall/precision“recall” and “precision”
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved.wikipedia
85 Related Articles

Information retrieval

queryretrievalqueries
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved. In information retrieval contexts, precision and recall are defined in terms of a set of retrieved documents (e.g. the list of documents produced by a web search engine for a query) and a set of relevant documents (e.g. the list of all documents on the internet that are relevant for a certain topic), cf. In the field of information retrieval, precision is the fraction of retrieved documents that are relevant to the query:
Traditional evaluation metrics, designed for Boolean retrieval or top-k retrieval, include precision and recall.

Positive and negative predictive values

positive predictive valuenegative predictive valuepositive
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved.
In information retrieval, the PPV statistic is often called the precision.

Relevance

irrelevantrelevantrelevancy
Both precision and recall are therefore based on an understanding and measure of relevance.
Given a conception of relevance, two measures have been applied: Precision and recall:

Statistical classification

classificationclassifierclassifiers
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved.
The measures precision and recall are popular metrics used to evaluate the quality of a classification system.

Receiver operating characteristic

ROC curveAUCROC
Recall and Inverse Recall, or equivalently true positive rate and false positive rate, are frequently plotted against each other as ROC curves and provide a principled mechanism to explore operating point tradeoffs.
The true-positive rate is also known as sensitivity, recall or probability of detection in machine learning.

Youden's J statistic

InformednessYouden's index
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).
An unrelated but commonly used combination of basic statistics from Information Retrieval is the F-score, being a (possibly weighted) harmonic mean of recall and precision where recall = sensitivity = true positive rate, but specificity and precision are totally different measures.

Harmonic mean

weighted harmonic meanharmonicHarmonically
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).
In computer science, specifically information retrieval and machine learning, the harmonic mean of the precision (true positives per predicted positive) and the recall (true positives per real positive) is often used as an aggregated performance score for the evaluation of algorithms and systems: the F-score (or F-measure).

F1 score

F-MeasureF-scoreF1
The two measures are sometimes used together in the F1 Score (or f-measure) to provide a single measurement for a system.
It considers both the precision p and the recall r of the test to compute the score: p is the number of correct positive results divided by the number of all positive results returned by the classifier, and r is the number of correct positive results divided by the number of all relevant samples (all samples that should have been identified as positive).

Type I and type II errors

Type I errorfalse-positivefalse positive
In statistics, if the null hypothesis is that all items are irrelevant (where the hypothesis is accepted or rejected based on the number selected compared with the sample size), absence of type I and type II errors (i.e. perfect sensitivity and specificity of 100% each) corresponds respectively to perfect precision (no false positive) and perfect recall (no false negative).

Accuracy and precision

accuracyprecisionaccurate
Accuracy is a weighted arithmetic mean of Precision and Inverse Precision (weighted by Bias) as well as a weighted arithmetic mean of Recall and Inverse Recall (weighted by Prevalence). The first problem is 'solved' by using Accuracy and the second problem is 'solved' by discounting the chance component and renormalizing to Cohen's kappa, but this no longer affords the opportunity to explore tradeoffs graphically.
Commonly used metrics include the notions of precision and recall.

Sensitivity and specificity

sensitivityspecificitysensitive
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved. In statistics, if the null hypothesis is that all items are irrelevant (where the hypothesis is accepted or rejected based on the number selected compared with the sample size), absence of type I and type II errors (i.e. perfect sensitivity and specificity of 100% each) corresponds respectively to perfect precision (no false positive) and perfect recall (no false negative).
The values of sensitivity and specificity are agnostic to the percent of positive cases in the population of interest (as opposed to, for example, precision).

Uncertainty coefficient

proficiencyProficiency (metric)Theil's U
The uncertainty coefficient is useful for measuring the validity of a statistical classification algorithm and has the advantage over simpler accuracy measures such as precision and recall in that it is not affected by the relative fractions of the different classes, i.e., P(x)

Pattern recognition

pattern analysispattern detectionpatterns
In pattern recognition, information retrieval and classification (machine learning), precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of the total amount of relevant instances that were actually retrieved.

Search engine (computing)

search enginesearch enginessearch engines
When a search engine returns 30 pages only 20 of which were relevant while failing to return 40 additional relevant pages, its precision is 20/30 = 2/3 while its recall is 20/60 = 1/3.

Statistics

statisticalstatistical analysisstatistician
In statistics, if the null hypothesis is that all items are irrelevant (where the hypothesis is accepted or rejected based on the number selected compared with the sample size), absence of type I and type II errors (i.e. perfect sensitivity and specificity of 100% each) corresponds respectively to perfect precision (no false positive) and perfect recall (no false negative).

Null hypothesis

nullnull hypotheseshypothesis
In statistics, if the null hypothesis is that all items are irrelevant (where the hypothesis is accepted or rejected based on the number selected compared with the sample size), absence of type I and type II errors (i.e. perfect sensitivity and specificity of 100% each) corresponds respectively to perfect precision (no false positive) and perfect recall (no false negative).

Matthews correlation coefficient

Matthews Correlation
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).

Geometric mean

geometric averagegeometricmean
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).

Linear regression

regression coefficientmultiple linear regressionregression
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).

Markedness

unmarkedmarked(unmarked)
Examples of measures that are a combination of precision and recall are the F-measure (the weighted harmonic mean of precision and recall), or the Matthews correlation coefficient, which is a geometric mean of the chance-corrected variants: the regression coefficients Informedness (DeltaP') and Markedness (DeltaP).

Cohen's kappa

kappakappa statisticKappa coefficient
The first problem is 'solved' by using Accuracy and the second problem is 'solved' by discounting the chance component and renormalizing to Cohen's kappa, but this no longer affords the opportunity to explore tradeoffs graphically.

Web search engine

search enginesearch enginesweb search
In information retrieval contexts, precision and recall are defined in terms of a set of retrieved documents (e.g. the list of documents produced by a web search engine for a query) and a set of relevant documents (e.g. the list of all documents on the internet that are relevant for a certain topic), cf.

Relevance (information retrieval)

relevancerelevantrelevancy ranking
In the field of information retrieval, precision is the fraction of retrieved documents that are relevant to the query:

Contingency table

cross tabulationcontingency tablescrosstab
The four outcomes can be formulated in a 2×2 contingency table or confusion matrix, as follows:

Confusion matrix

confusion matrices
The four outcomes can be formulated in a 2×2 contingency table or confusion matrix, as follows: