AWS Certified Machine Learning – Specialty — Question 142

A company that manufactures mobile devices wants to determine and calibrate the appropriate sales price for its devices. The company is collecting the relevant data and is determining data features that it can use to train machine learning (ML) models. There are more than 1,000 features, and the company wants to determine the primary features that contribute to the sales price.
Which techniques should the company use for feature selection? (Choose three.)

Answer options

• A. Data scaling with standardization and normalization
• B. Correlation plot with heat maps
• C. Data binning
• D. Univariate selection
• E. Feature importance with a tree-based classifier
• F. Data augmentation

Correct answer: B, D, E

Explanation

The correct techniques for feature selection are B, D, and E. A correlation plot with heat maps (B) helps visualize relationships between features and the sales price, while univariate selection (D) evaluates each feature individually for its effect on the target variable. Feature importance with a tree-based classifier (E) identifies which features provide the most predictive power. Options A, C, and F do not focus on feature selection directly, making them unsuitable for the company's goal.