Classification Metrics
17 practice items · needed by 3 roles
Easy 7Medium 7Hard 3
Practice questions
- Definition of precisioneasy
- Accuracy on imbalanced dataeasy
- Recall from confusion countseasy
- F1 score from a confusion matrixmedium
- Log loss of a poor predictionmedium
- AUC from four scoresmedium
- ROC-AUC after calibrationhard
- Precision from a confusion matrixeasy
- Metric when missing a case is costlyeasy
- High accuracy on rare eventseasy
- What an AUC of 0.8 meansmedium
- F1 from precision and recallmedium
- Which prediction hurts log loss mostmedium
- Brier score for two predictionsmedium
- AUC after squaring the scoreshard
- Precision when prevalence dropshard
- Recall from prediction countseasy