Accuracy Calculator

Your details

Choose "confusion matrix" for machine-learning or diagnostic-test results. Use "prevalence-adjusted" when your sample does not reflect real-world disease prevalence. Use "percent error" to compare a measured value with a true reference value.
Cases that are actually positive and were correctly predicted as positive.
Cases that are actually negative and were correctly predicted as negative.
Cases that are actually negative but were incorrectly predicted as positive. Also called a Type I error.
Cases that are actually positive but were incorrectly predicted as negative. Also called a Type II error.
AccuracyGood accuracy
94%

Proportion of all predictions that were correct

Precision (PPV)75%
Recall (Sensitivity)75%
F1 Score75%
Specificity (TNR)96.59%
False Positive Rate3.41%
Prevalence12%
Total cases1,000
94% %
Poor<70Fair70-85Good85-95Excellent95+

Overall accuracy is 94.0%.

  • The model correctly classified 94.0% of 1,000 cases.
  • Precision (75.0%) and recall (75.0%) are well balanced.
  • The F1 score is 75.0%, the harmonic balance between precision and recall.

Next stepSpecificity is 96.6% and the false-positive rate is 3.4%. Consider plotting the ROC curve (specificity vs. recall across thresholds) to see the full trade-off.

 = Powered by OnlyCalculators