Precision is the fraction of positive predictions that are actually correct: TP / (TP + FP). High precision means few false alarms. It is important when the cost of a false positive is high, such as in spam filtering.

What is recall (sensitivity)?

Recall measures how many actual positives were correctly identified: TP / (TP + FN). High recall means few missed positives. It matters when missing a true positive is costly, such as in medical diagnosis.

What is the F1 score?

The F1 score is the harmonic mean of precision and recall: 2 * (Precision * Recall) / (Precision + Recall). It balances both metrics and is especially useful for imbalanced class distributions.

When should I use F1 score vs. accuracy?

Accuracy is misleading when classes are imbalanced — a model predicting all negatives scores high accuracy on a 99% negative dataset but has zero recall. F1 score accounts for false negatives and false positives, making it a better metric for imbalanced problems.

Specificity (True Negative Rate) is TN / (TN + FP). It measures how well the model identifies negatives. It is the complement of the false positive rate and is used in medical testing alongside sensitivity (recall).

What is the Matthews Correlation Coefficient (MCC)?

MCC is a single metric that accounts for all four confusion matrix values. It ranges from -1 (total disagreement) to +1 (perfect prediction) with 0 indicating random chance. It is considered one of the most informative single scores for binary classification.

Precision Recall Calculator

Enter TP, FP, TN, FN from your confusion matrix and instantly get precision, recall, F1 score, accuracy, and more.

Confusion Matrix Inputs

Enter the four values from your binary classifier's confusion matrix.

True Positives (TP)

Correctly predicted positive

False Positives (FP)

Negative predicted as positive

False Negatives (FN)

Positive predicted as negative

True Negatives (TN)

Correctly predicted negative

Confusion Matrix

	Predicted Positive	Predicted Negative
Actual Positive	90	5
Actual Negative	10	795

Classification Metrics

N = 900

F-Beta Score

Adjust beta to weight recall (beta > 1) or precision (beta < 1) more heavily.

Beta (β)

F-Beta = 0.9231

Copied!

Summary

This tool computes the core binary classification metrics — precision, recall, F1 score, accuracy, and specificity — from the four values of a confusion matrix. It accepts True Positives (TP), False Positives (FP), True Negatives (TN), and False Negatives (FN) and derives all metrics client-side with no data sent to any server. Results update live as you type.

How it works

Enter the number of True Positives (TP) — correctly predicted positive cases.
Enter the number of False Positives (FP) — negative cases incorrectly predicted as positive.
Enter the number of True Negatives (TN) — correctly predicted negative cases.
Enter the number of False Negatives (FN) — positive cases incorrectly predicted as negative.
All classification metrics are computed instantly in your browser.
Use the copy button next to any metric to copy its value to the clipboard.

Use cases

Evaluate the performance of a binary classification ML model.
Compare precision vs. recall trade-offs when tuning a decision threshold.
Report F1 score for imbalanced datasets where accuracy is misleading.
Verify hand-calculated confusion matrix metrics during coursework.
Assess spam filter performance using TP/FP/TN/FN counts.
Benchmark object detection or anomaly detection models.
Debug model results by checking which metric is underperforming.
Compute Matthews Correlation Coefficient (MCC) for balanced evaluation.

Frequently Asked Questions

Last updated: 2026-05-23 · Reviewed by Nham Vu