AIF-C01 Question 39
Single answerA data scientist is tasked with building a machine learning model to predict customer churn for an e-commerce platform. She decides to follow the ML development lifecycle. At which stage should she focus on selecting an appropriate algorithm and tuning its hyperparameters?
- A
Data Collection and Preparation
- B
Model Training
- C
Model Evaluation
- D
Model Deployment
Show answer and explanation
Correct answer: B
Explanation
The machine learning development lifecycle consists of distinct stages: data collection and preparation, model training, model evaluation, and model deployment. During the model training stage, the focus is on selecting the best algorithm and tuning its hyperparameters to achieve optimal model performance. This makes it the correct stage for the described task.
- A. Incorrect.
Data Collection and Preparation involves gathering, cleaning, and preprocessing the data. Algorithm selection and hyperparameter tuning are not performed in this stage.
- B. Correct.
Model Training involves selecting an appropriate algorithm, feeding the training data into the algorithm, and tuning its hyperparameters to optimize model performance. This is the correct stage for the described task.
- C. Incorrect.
Model Evaluation focuses on assessing the model's performance on test data. It does not involve selecting algorithms or hyperparameter tuning.
- D. Incorrect.
Model Deployment involves integrating the trained and evaluated model into a production environment. Algorithm selection and hyperparameter tuning are completed prior to this stage.