AIF-C01 Question 41
Select 2You are building an end-to-end machine learning pipeline to predict customer churn for a subscription service. After collecting and cleaning the data, you want to ensure that your pipeline can transform raw data into meaningful features and evaluate the model's performance. Which steps in the ML pipeline are essential for this task?
- A
Feature engineering
- B
Model training
- C
Hyperparameter tuning
- D
Evaluation
- E
Monitoring
Show answer and explanation
Correct answers: A, D
Explanation
In an ML pipeline, feature engineering transforms raw data into features that the model can use, while evaluation measures the model's performance to ensure it meets business objectives. These steps are critical to preparing the data and assessing the model's predictive capabilities before deployment.
- A. Correct.
Feature engineering is essential for transforming raw data into meaningful features that the model can interpret effectively. This step ensures the data is prepared for the model.
- B. Incorrect.
Model training involves fitting the machine learning algorithm to the data, but it is not directly involved in transforming raw data or evaluating the model's performance.
- C. Incorrect.
Hyperparameter tuning is focused on optimizing model parameters for better results, but it does not directly transform raw data or evaluate performance.
- D. Correct.
Evaluation is necessary to measure the model's performance using metrics such as accuracy, precision, or recall. This step ensures the model is effective in predicting customer churn.
- E. Incorrect.
Monitoring is important for tracking a deployed model's performance in production, but it is not part of the pipeline when transforming raw data or evaluating the model.