Recall is... #flashcard
- The percentage of actual positives that the model correctly identifies.
- Intuition: "What percentage of all harmful content does the model catch?"
- Calculation: (True Positives) / (True Positives + False Negatives)
- Situation: Use when the cost of false negatives is high.
- Avoid: When you need to account for false positives as well.
<!--ID: 1751507777219-->