4. Precision and Recall
Precision: % of selected items that are correct tp/(tp+fp) (slected items must be correct)
Recall: % of correct items that are selected tp/(tp + fn) (all correct items are selected)
F1 Score: 2PR/(P+R)
Accuracy (ACC) = Σ True positive + Σ True negative/Σ Total population
Misclassification error: 1-accuracy
Trade-off
- Namely, that you can often increase the precision of a classifier but the downside is that you may reduce recall, or you could increase the recall of a classifier at the cost of reducing precision.
Recall oriented machine learning tasks:
- include medical and legal applications, where the consequences of not correctly identifying a positive example can be high.
precision-oriented tasks:
- Many customer-facing machine learning tasks
- search engine ranking, query suggestion, documents classification
- the consequences of false positives can be high, for example, hurting the customer's experience on a website by providing incorrect or unhelpful information.