Classification performance is derived from a confusion matrix that counts true positives, false positives, true negatives, and false negatives, and several metrics answer different questions from those counts. Recall, also called sensitivity or the true positive rate, is calculated as true positives divided by the sum of true positives and false negatives. It answers the question of what share of all genuinely positive cases the model managed to find, which makes option B correct. Recall is the priority where missing a positive is costly, such as screening for disease or detecting fraud. Precision, option A, divides true positives by the sum of true positives and false positives; it measures how many of the cases the model flagged as positive are truly positive, and it matters when false alarms are expensive. Specificity, option C, is the true negative rate, the proportion of actual negatives correctly identified. Precision and recall typically move in opposite directions as the decision threshold changes, and the F1 score combines them into a single value through the harmonic mean. Selecting the metric that matches the business consequence of each error type is a key part of model evaluation. Reference: Data Science Fundamentals Study Guide – Data Science Process: Model Evaluation and Performance Metrics.
===============
Contribute your Thoughts:
Chosen Answer:
This is a voting comment (?). You can switch to a simple comment. It is better to Upvote an existing comment if you don't have anything to add.
Submit