Recall

What is Recall?

In the field of artificial intelligence, recall is a key evaluation metric used to assess the performance of a machine learning model, particularly in classification tasks. It quantifies the model’s effectiveness in identifying all true positive instances from the dataset. Recall is calculated as the ratio of true positives to the sum of true positives and false negatives. High recall indicates that the model is able to identify most of the relevant examples, which is crucial in scenarios where missing an important instance can have significant consequences. For instance, in medical diagnostics, a high recall model ensures that most diseased patients are correctly identified, minimizing the risk of missing a diagnosis. However, high recall may sometimes come at the expense of precision, leading to more false positives. Therefore, it is often balanced with precision to provide a more comprehensive understanding of a model's performance through metrics like the F1 score.

Recall is the measure of a model’s ability to capture all relevant instances in a dataset.

Examples

• Medical Diagnosis: In a cancer detection system, high recall ensures that most cancer cases are identified, even if it means flagging some non-cancerous cases for further testing.


• Spam Detection: In email filtering systems, high recall helps in catching most spam emails, ensuring that users are not exposed to harmful or unwanted content.

Additional Information

• Recall is also known as sensitivity or true positive rate.


• It is particularly important in applications where missing a relevant instance has serious consequences.