Recall

As explained in Machine Learning Algorithms in Depth

The proportion of actual positive examples correctly predicted as positive.

Recall is the proportion of actual positive examples predicted as positive. For a binary classifier, recall, also called true positive rate (TPR), is Recall = TP/(TP + FN), where TP counts true positives and FN counts false negatives. section 5.2

Use recall to interpret classifier performance:

• A higher recall means fewer actual positives are missed as false negatives. section 5.2
• Changing the classification threshold changes TP, FN, and therefore recall. section 5.2
• Precision and recall together form a precision-recall curve; their fixed-threshold harmonic mean is the F1 score. section 5.2

Figure 5.4 Receiver operating characteristic (ROC) plot (left) and precision-recall plot (right)

Compare precision and recall across classification thresholds with a precision-recall plot. section 5.2

How Machine Learning Algorithms in Depth covers it

How deep it goes

Defines recall algebraically and relates it to thresholds, confusion matrices, precision-recall curves, and F1, without deriving implementation details.

Examples

Uses TP and FN counts, threshold variation, and precision-recall plots as concrete interpretive examples.

What it assumes

Assumes binary classification and confusion-matrix terminology while defining recall from true-positive and false-negative counts.

What you can do after

Evaluate binary classifiers by calculating recall and interpreting its tradeoff with precision across thresholds.

How it connects

Connects recall with true positive rate, confusion matrices, false negatives, precision, classification thresholds, precision-recall curves, and F1 score.

liveBookmanning.com

Read Machine Learning Algorithms in Depth on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Grokking Machine LearningA classification metric equal to the proportion of positive-labeled data points that are correctly predicted as positive.
• 

  liveBookmanning.com
  Starting Data Analytics with Generative AI and PythonThe proportion of actual positive cases correctly classified as positive, equivalent to sensitivity.
• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionRecall is the proportion of relevant items that a retrieval or classification system successfully identifies.
• 

  liveBookmanning.com
  Machine Learning System DesignThe proportion of actual positive cases that a model correctly predicts as positive.
• 

  liveBookmanning.com
  Effective Conversational AIRecall is the proportion of relevant or correct cases that a classifier successfully identifies.
• 

  liveBookmanning.com
  Delivering Machine Learning Projects: From design to deploymentRecall is the proportion of relevant positive instances that a predictive classification system correctly identifies.
• 

  liveBookmanning.com
  Statistics Slam DunkSensitivity is the proportion of actual positive cases that a classification procedure correctly identifies.
• 

  liveBookmanning.com
  Knowledge Graphs and LLMs in ActionRecall is the proportion of relevant positive instances that are successfully identified.
• 

  liveBookmanning.com
  Deep Learning with PyTorch, Second EditionA classification metric measuring the fraction of actual positives that are correctly identified, which can be undefined when there are no actual positives.
• 

  liveBookmanning.com
  Grokking StatisticsTrue positive rate is the proportion of positive instances correctly classified as positive.
• 

  liveBookmanning.com
  Statistics Every Programmer NeedsSensitivity is the proportion of actual positive instances that a classifier correctly identifies as positive.
• 

  liveBookmanning.com
  Just Use Postgres!Recall is the proportion of relevant items that a search system retrieves from all relevant items available.
• 

  liveBookmanning.com
  The Art of AI Product DevelopmentRecall is the proportion of actual positive cases that a classifier correctly identifies.
• 

  liveBookmanning.com
  Generative AI in ActionRecall is the proportion of relevant or actual positive cases that a system correctly identifies.
• 

  liveBookmanning.com
  Financial AI in PracticeThe proportion of all fraudulent cases found within a selected group of model-ranked cases, such as the top n percent.
• 

  liveBookmanning.com
  Data-Driven CybersecurityRecall is the proportion of actual instances in a class that a classification model correctly identifies.
• 

  liveBookmanning.com
  Ensemble Methods for Machine LearningRecall is the fraction of actual positive examples that are correctly predicted as positive.
• 

  liveBookmanning.com
  LLMs in ProductionRecall is the proportion of reference items that are correctly identified by a prediction.
• 

  liveBookmanning.com
  Introduction to Generative AI, Second EditionRecall is the proportion of actual positive cases that a classifier correctly identifies.
• 

  liveBookmanning.com
  The AI Pocket BookA classification or retrieval metric measuring how many relevant or positive items are successfully found.
• 

  liveBookmanning.com
  Elasticsearch in Action, Second EditionIn information retrieval, recall is the proportion of all relevant documents that are retrieved.
• 

  liveBookmanning.com
  Time Series Forecasting Using Foundation ModelsRecall is the proportion of actual positive observations that a detection system correctly identifies.
• 

  liveBookmanning.com
  Fabulous Adventures in Data Structures and AlgorithmsTest sensitivity is the conditional probability that a medical test is positive when the tested condition is present.

Compare how 24 books explain recall →

Related concepts

• f-scoreused by · An F-score is a classification or evaluation metric combining precision and recall via their weighted harmonic mean; F1 is the equal-weight case.
• precisioncontrasts with · Precision is the proportion of items identified as positive, selected, or retrieved that satisfy the relevant correctness or relevance criterion.
• false-negative ratecontrasts with · The false-negative rate is the proportion of actual positive cases that are incorrectly classified or identified as negative.
• receiver operating characteristicused by · A receiver operating characteristic curve plots a binary classifier’s true-positive rate against its false-positive rate as the decision threshold varies.
• similarity searchmeasures · Similarity search is a retrieval technique that finds or ranks items by comparing their representations with a query according to a similarity or distance measure.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning Algorithms in Depth. The section links point to the parts of the book it draws on.