Precision-recall curve

As explained in Machine Learning Algorithms in Depth

A precision-recall curve plots precision against recall across classification thresholds.

A precision–recall curve evaluates a binary classifier across classification thresholds by plotting precision against recall. Each threshold changes the confusion-matrix counts and therefore produces a different point. section 5.2

Precision measures the fraction of predicted positives that are correct: TP/(TP + FP). Recall measures the fraction of actual positives detected: TP/(TP + FN). Averaging precision over recall values approximates the curve’s area. section 5.2

Threshold comparison

Adjusting a classifier’s threshold traces different precision–recall points. For a fixed threshold, the F1 score summarizes both measures as 2PR/(P + R), their harmonic mean. section 5.2

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

Compare the precision–recall plot with the receiver operating characteristic plot in the paired figure. section 5.2

How Machine Learning Algorithms in Depth covers it

How deep it goes

Defines both axes, relates them to confusion-matrix counts, and adds mean precision and F1 summaries.

Examples

Uses a precision–recall plot and contrasts it visually with a receiver operating characteristic plot.

What it assumes

Assumes binary-classification thresholds and confusion-matrix counts including true positives, false positives, and false negatives.

What you can do after

Evaluate classifier performance across thresholds and summarize it with mean precision or F1.

How it connects

Connects classification thresholds with confusion matrices, receiver operating characteristic curves, area under the curve, and F1 scores.

liveBookmanning.com

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

Also explained in

• 

  liveBookmanning.com
  Grokking StatisticsA precision-recall curve plots a classifier's precision against its recall across decision thresholds.

Compare how 2 books explain precision-recall curve →

Related concepts

• area under the curvemeasured by · The area under a curve is the integral of its values over a specified domain and, in classification evaluation, commonly summarizes performance across decision thresholds.
• precisionuses · Precision is the proportion of items identified as positive, selected, or retrieved that satisfy the relevant correctness or relevance criterion.
• recalluses · Recall is the fraction of relevant, positive, or target outcomes that a system successfully identifies, retrieves, or captures.
• receiver operating characteristicalternative to · A receiver operating characteristic curve plots a binary classifier’s true-positive rate against its false-positive rate as the decision threshold varies.
• class imbalanceused by · Class imbalance is a condition in a classification dataset where classes have substantially unequal numbers of examples.
• fraud detectionmeasures · Fraud detection is the process of identifying transactions, activities, or other events suspected of involving intentional deception or unauthorized gain.

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.