Google Professional Data Engineer

Get started today

Ultimate access to all questions.

Explanation:

The correct answer is B. Expanding the training dataset exposes the model to a wider variety of examples, enhancing its ability to generalize and reducing overfitting. Simultaneously, reducing the number of input features simplifies the model, helping it focus on the most relevant patterns rather than memorizing noise.

Why not A? While increasing the dataset size is beneficial, adding more features can make the model overly complex, worsening overfitting.
Why not C? Reducing the dataset size while adding features can lead to underfitting, where the model fails to capture essential data patterns.
Why not D? Decreasing both the dataset size and feature count can severely limit the model's learning capability, leading to poor performance.

Explanation:

Why not A? While increasing the dataset size is beneficial, adding more features can make the model overly complex, worsening overfitting.
Why not C? Reducing the dataset size while adding features can lead to underfitting, where the model fails to capture essential data patterns.
Why not D? Decreasing both the dataset size and feature count can severely limit the model's learning capability, leading to poor performance.

Comments (0)

No comments yet.

You're developing a deep learning model to predict the likelihood of a customer making a purchase on your ecommerce site. Initial evaluations show the model is overfitting with both training and test data. What steps can you take to improve the model's accuracy on new data?

Real Exam

Expand the training dataset and add more input features.

5.9%

Expand the training dataset and reduce the number of input features.

82.4%

Shrink the training dataset and add more input features.

5.9%

Shrink the training dataset and reduce the number of input features.

5.9%