A data engineer is evaluating customer data in Amazon SageMaker Data Wrangler. The data engineer will use the customer data to create a new model to predict customer behavior. The engineer needs to…

Question

Options

• AUse SageMaker Data Wrangler to refit and transform the dataset by applying one-hot encoding to
• BUse SageMaker Data Wrangler diagnostic visualization. Use principal components analysis
• CUse the SageMaker Data Wrangler Quick Model visualization to quickly evaluate the dataset and
• DUse the SageMaker Data Wrangler Min Max Scaler transform to normalize the data.
• EUse SageMaker Data Wrangler diagnostic visualization. Use least absolute shrinkage and

• A
  31% (9)
• B
  48% (14)
• C
  7% (2)
• D
  14% (4)

Explanation

Multicollinearity occurs when two or more features are highly correlated, which can degrade model performance. SageMaker Data Wrangler provides built-in diagnostic visualizations that can detect this with no custom code. Option B uses Principal Component Analysis (PCA), which identifies directions of maximum variance and reveals correlated feature groups - a standard technique for diagnosing multicollinearity. Option E uses LASSO (Least Absolute Shrinkage and Selection Operator), which applies L1 regularization to shrink coefficients of correlated/redundant features toward zero, effectively flagging them. Both are native Data Wrangler diagnostics requiring minimal effort. One-hot encoding (A) handles categorical variables, not correlation. Quick Model (C) provides a general accuracy estimate, not a multicollinearity analysis. Min-Max Scaler (D) normalizes feature ranges but does not address feature correlation.

No community discussion yet for this question.