Opening your learning space
Starting the app …
Opening your learning space
Starting the app …
Recall is the fraction of actual positive cases detected by a classifier.
Free to browse · flashcards or a multiple-choice quiz · import from Essential
Formula
dimensionless, between 0 and 1; convention-dependent when TP+FN=0
| Symbol | Meaning | Unit |
|---|---|---|
| True positive predictions | count | |
| False negative cases | count |
\mathrm{Recall}=\frac{TP}{TP+FN}Recall is the sensitivity of the specified positive class. It asks what fraction of all actual positive cases were found and ignores false positives and true negatives. The value depends on the decision threshold and test data; lowering a threshold often raises recall but also increases false alarms. If the test set contains no positive cases, the denominator is zero and recall is undefined. State the class and threshold with the score.
Recall is the sensitivity of the specified positive class. It asks what fraction of all actual positive cases were found and ignores false positives and true negatives. The value depends on the decision threshold and test data; lowering a threshold often raises recall but also increases false alarms. If the test set contains no positive cases, the denominator is zero and recall is undefined. State the class and threshold with the score.
Recall is the fraction of actual positive cases detected by a classifier.
False negatives from recall
Recall lies between 0 and 1 when its denominator is positive.
Use the example for Recall (sensitivity) to explain which inputs determine the requested output.
The example connects concrete inputs to its result.
Recall is the fraction of actual positive cases detected by a classifier.
The relation applies only under its stated assumptions.
Applying the relation to a different case without checking its stated assumptions.
Check the validity range before substituting values: Recall is the sensitivity of the specified positive class. It asks what fraction of all actual positive cases were found and ignores false positives and true negatives. The value depends.
Formula cluster
Recall is the fraction of actual positive cases detected by a classifier.
Of 50 actually relevant cases, a model finds TP=40 and misses FN=10. Recall=40/(40+10)=0.80, so it detects four fifths of the positive cases.
Measure completeness in search, screening or alerting when missed positive cases are especially costly.
Here is how to work through a typical Recall (sensitivity) (Recall = TP/(TP+FN)) task step by step:
Task
Use the example for Recall (sensitivity) to explain which inputs determine the requested output.
Solution path
The example connects concrete inputs to its result.
Recall is the fraction of actual positive cases detected by a classifier.
The relation applies only under its stated assumptions.
Recall is the fraction of actual positive cases identified as positive. With TP=40 and FN=10, the model finds 40 of 50 positive cases, so recall is 0.80. False positive alerts are not in its denominator. It measures missed cases, not the reliability of positive predictions; precision addresses that second question. Recall measures completeness for a specified positive class and depends on threshold and test composition. Compare the same target class under documented evaluation conditions. Sampling variation should also be assessed using case counts and uncertainty intervals. This keeps the evaluation interpretable.
Read TP as correctly detected positive cases and FN as positive cases the model missed. Compute TP/(TP+FN). For TP=40 and FN=10, recall is 40/50=0.80. Check whether matrix rows or columns represent actual classes; conventions differ. Misreading a cell can swap recall with precision. Recall measures completeness for a specified positive class and depends on threshold and test composition. Compare the same target class under documented evaluation conditions. Sampling variation should also be assessed using case counts and uncertainty intervals. This keeps the evaluation interpretable.
Lowering a decision threshold usually classifies more cases as positive. It can find additional actual positives, reduce FN and raise recall. At the same time FP often increases, which may lower precision. This trade-off is a property of the decision rule, not a calculation error. Report the threshold and inspect multiple values with a precision-recall curve. Recall measures completeness for a specified positive class and depends on threshold and test composition. Compare the same target class under documented evaluation conditions. Sampling variation should also be assessed using case counts and uncertainty intervals. This keeps the evaluation interpretable.
For binary classification of the chosen positive class, recall, sensitivity and true positive rate are the same quantity: TP/(TP+FN). The names come from different application areas. In multiclass problems, recall is computed per class, and macro, micro or weighted averaging combines values differently. State the aggregation rather than reporting an unspecified overall recall. Recall measures completeness for a specified positive class and depends on threshold and test composition. Compare the same target class under documented evaluation conditions. Sampling variation should also be assessed using case counts and uncertainty intervals. This keeps the evaluation interpretable.
If TP+FN=0, the test set contains no actual positive cases. Recall is undefined because there is no positive case whose detection can be measured. A library may warn or insert a fallback value, but that does not improve the evaluation. Ensure the test includes representative positives or report the metric as unavailable. Recall measures completeness for a specified positive class and depends on threshold and test composition. Compare the same target class under documented evaluation conditions. Sampling variation should also be assessed using case counts and uncertainty intervals. This keeps the evaluation interpretable.
Quanta exam set
Curated exam set for "Recall (sensitivity)":
Question (front)
Which formula describes Recall (sensitivity)?
Answer in your set
Question (front)
How do you rearrange Recall (sensitivity) for False negatives from recall?
Answer in your set
Question (front)
Which common mistake happens with Recall (sensitivity)?
Answer in your set
+ 7 more cards: units, variables, derivation, example, exam task
The formula set contains 10 cards. After importing, review them and start practicing; your ratings determine later review intervals.
This list names subject references and further reading. A page or section number appears when it is part of the reference; these are not verbatim quotations.
Create a curated FSRS exam set for Recall = TP/(TP+FN): formula recall, variables, derivation, rearrangement, worked example, common mistakes and exam context.
Curated formula set · LaTeX · FSRS spaced repetition · import from Essential