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
Recall oriented machine learning tasks:
precision-oriented tasks: