Precision vs recall — when do you optimize each?
easyAnswer
- Precision = TP / (TP + FP): of the items flagged positive, how many really are.
- Recall = TP / (TP + FN): of the actual positives, how many we caught.
- Optimize precision when false positives are costly (spam filter, fraud alert to a user).
- Optimize recall when missing a positive is costly (cancer screening, security threats).
- F1 balances them.
Check yourself — multiple choice
- Precision measures how many positives we found; recall measures purity
- Precision = TP/(TP+FP), Recall = TP/(TP+FN)
- Precision and recall are the same on balanced data
- Recall is always more important than precision
Precision cares about the purity of positive predictions; recall cares about coverage of true positives.
#metrics#classification
Practise Supervised Learning
215 interview questions in this topic.
Related questions
- How do you read a confusion matrix and derive the key metrics?
- Why is accuracy a bad primary metric for many real-world problems?
- What is F-beta score and when is F1 not enough?
- Macro vs micro vs weighted averaging in multi-class metrics — how do you choose?
- What is Cohen's kappa and why is it useful?
- What is the Matthews Correlation Coefficient (MCC) and when should you use it?