MLS-C01 exam dumps

MLS-C01 practice question 229 of 389

AWS Certified Machine Learning - Specialty. Expert level, Amazon Web Services. Free question with the correct answer and a full explanation.

MLS-C01 Question 229

Single answer

You are developing a binary classification model to predict whether a customer will churn or not. During the evaluation phase, the business stakeholders indicate that minimizing false negatives (i.e., predicting a customer will not churn when they actually do) is critical as it leads to a loss in revenue. Which evaluation metric should you prioritize to align with the business requirement?

  1. A

    Accuracy

  2. B

    Precision

  3. C

    Recall

  4. D

    Root Mean Square Error (RMSE)

Show answer and explanation

Correct answer: C

Explanation

To meet the business requirement of minimizing false negatives, recall is the appropriate metric to prioritize. Recall focuses on capturing as many true positives as possible while minimizing the false negatives, which aligns with the need to identify all customers who are at risk of churning. Other metrics like accuracy and precision do not directly address this requirement, and RMSE is not applicable for classification problems.

  • A. Incorrect.

    Accuracy measures the overall correctness of the model by dividing the number of correct predictions by the total predictions. However, it does not specifically focus on false negatives, which is the business's primary concern.

  • B. Incorrect.

    Precision measures the proportion of true positives out of all predicted positives, focusing on reducing false positives. It is not the best metric when minimizing false negatives is the main priority.

  • C. Correct.

    Recall measures the proportion of true positives out of all actual positives, focusing on minimizing false negatives. This makes recall the correct metric to prioritize in this scenario.

  • D. Incorrect.

    Root Mean Square Error (RMSE) is a regression metric that measures the average magnitude of prediction errors in a continuous output. It is irrelevant for binary classification tasks.

Timed practice exam

Take a MLS-C01 practice test under exam conditions

65 questions in 180 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam