AI-900 Question 119
Single answerYou are building a machine learning model to predict customer churn for a subscription service. You divide your dataset into training and validation subsets. Which of the following best describes the purpose of the training and validation datasets?
- A
The training dataset is used to evaluate the model's performance, while the validation dataset is used to tune hyperparameters.
- B
The training dataset is used to fit the model to the data, while the validation dataset is used to evaluate the model's performance on unseen data.
- C
Both the training and validation datasets are used to adjust the weights of the model during training.
- D
The training dataset is used to preprocess the data, while the validation dataset is used to build the model.
Show answer and explanation
Correct answer: B
Explanation
In supervised machine learning, the training dataset is used to fit the model (learn patterns and relationships from the data), while the validation dataset is used to evaluate how well the model generalizes to unseen data. This process helps detect overfitting and allows for hyperparameter tuning without directly exposing the model to the test data.
- A. Incorrect.
This is incorrect because the training dataset is used to train the model (fit it to the data), not to evaluate its performance.
- B. Correct.
This is correct because the training dataset is used to train the model (fit the model to the data), and the validation dataset is used to evaluate the model's performance on unseen data to avoid overfitting.
- C. Incorrect.
This is incorrect because only the training dataset is used to adjust the model's weights during training, while the validation dataset is used for evaluation purposes.
- D. Incorrect.
This is incorrect because preprocessing can be applied to both datasets, but the training dataset is primarily used for training the model, and the validation dataset is used for evaluation.