Accuracy Metric

What is Accuracy?

Accuracy is the proportion of correct predictions (both true positives and true negatives) among the total number of cases examined.

Formula

Accuracy = (TP + TN) / (TP + TN + FP + FN)

Range

• 0% to 100%: Higher is better
• 100%: All predictions correct
• 50%: Random guessing (for balanced binary)

Limitations

• Misleading for Imbalanced Data: High accuracy possible by always predicting the majority class
• Example: 95% healthy, 5% disease -> Predicting "healthy" always gives 95% accuracy

When to Use

• Balanced datasets (roughly equal class sizes)
• When all types of errors are equally important

Related Terms

• F1 Score: Better for imbalanced data
• Precision: Focus on positive predictions
• Recall: Focus on finding all positives