Recall

Recall

Recall is a metric used in classification to measure how many of the actual positive instances were correctly identified by the model. It is also known as sensitivity or the true positive rate.

Definition

$\text{Recall}=\frac{TP}{TP+FN}$

Where:

• TP = True Positives – correctly predicted positive instances
• FN = False Negatives – actual positives incorrectly predicted as negative

Recall answers the question: "Of all actual positive cases, how many did we correctly identify?"

Simple Example

A medical test is used to detect cancer. There are 100 people with cancer:

• The test correctly identifies 90 as having cancer (TP = 90)
• It misses 10 people (FN = 10)

$\text{Recall}=\frac{90}{90+10}=\frac{90}{100}=0.9=90\mathrm{\%}$

This means the test correctly detects 90% of cancer cases.

When to Use Recall

Recall is crucial when missing a positive case has serious consequences.

Real-World Scenarios

• Cancer diagnosis: Missing a sick patient (false negative) is risky.
• Fraud detection: It's better to catch all suspicious activity even if some are false alarms.
• Fire alarms: Better to alert even for minor smoke than miss a real fire.

High vs Low Recall

• High Recall: Most actual positives are identified.
• Low Recall: Many positives are missed (false negatives).

Related Metrics

• Precision
• F1 Score
• Confusion Matrix
• Sensitivity and Specificity

SEO Keywords

recall in machine learning, true positive rate, sensitivity, recall formula, classification evaluation, medical test recall, fraud detection model