MLS-C01 Question 366
Single answerA data scientist is building a binary classification model to predict whether an email is spam or not spam. During the evaluation phase, they notice that the model has a high accuracy but performs poorly on identifying spam emails. Which metric should the data scientist prioritize to better evaluate the model's performance for identifying spam emails?
- A
Precision
- B
Recall
- C
F1 Score
- D
Mean Absolute Error (MAE)
Show answer and explanation
Correct answer: B
Explanation
In binary classification tasks, recall is the metric that evaluates the ability of the model to correctly identify all positive cases (in this scenario, spam emails). Since the primary concern is the poor identification of spam emails, recall is the most appropriate metric to prioritize.
- A. Incorrect.
Precision measures how many of the emails predicted as spam are actually spam, but it does not address the issue of missed spam emails, which is the concern here.
- B. Correct.
Recall measures the proportion of actual spam emails correctly identified by the model. Since identifying spam emails is the priority, recall is the most relevant metric.
- C. Incorrect.
The F1 Score is a harmonic mean of precision and recall. While it balances both metrics, the question specifically emphasizes identifying spam emails, making recall the better choice.
- D. Incorrect.
Mean Absolute Error (MAE) is a regression metric and is not applicable for evaluating binary classification models.