Question 20
Domain 3: Train and evaluate modelsYou are a senior data scientist in your company. You are evaluating a completed binary classification machine learning model. You need to use precision as the evaluation metric. Which visualization should you use?
Correct answer: E
Explanation
A Precision-Recall curve is used to evaluate a binary classifier when precision is the metric of interest because it shows the tradeoff between "precision" and "recall" across thresholds. It is especially useful for binary classification models where class imbalance can make accuracy misleading.
Why each option is right or wrong
A. Receiver Operating Characteristic (ROC) curve
B. Violin plot
C. Gradient descent
D. Scatter plot
E. Precision-Recall curve
A Precision-Recall curve is the standard threshold-based diagnostic for a binary classifier when the positive-class precision is the quantity being assessed, because it plots precision against recall for all decision thresholds rather than at a single cutoff. In contrast to ROC analysis, it is the more informative visualization when the positive class is relatively rare, since precision directly reflects false-positive burden and recall shows the corresponding sensitivity tradeoff.