nerdexam
Amazon

MLS-C01 · Question #335

A data scientist obtains a tabular dataset that contains 150 correlated features with different ranges to build a regression model. The data scientist needs to achieve more efficient model training…

The correct answer is B. Load the data into Amazon SageMaker Data Wrangler. Scale the data with a Min Max Scaler. Before applying PCA to a dataset with 150 correlated features having different ranges, it is crucial to scale the data to ensure PCA's effectiveness and minimize impact on model performance.

Data Engineering

Question

A data scientist obtains a tabular dataset that contains 150 correlated features with different ranges to build a regression model. The data scientist needs to achieve more efficient model training by implementing a solution that minimizes impact on the model's performance. The data scientist decides to perform a principal component analysis (PCA) preprocessing step to reduce the number of features to a smaller set of independent features before the data scientist uses the new features in the regression model. Which preprocessing step will meet these requirements?

Options

  • AUse the Amazon SageMaker built-in algorithm for PCA on the dataset to transform the data.
  • BLoad the data into Amazon SageMaker Data Wrangler. Scale the data with a Min Max Scaler
  • CReduce the dimensionality of the dataset by removing the features that have the highest
  • DReduce the dimensionality of the dataset by removing the features that have the lowest

How the community answered

(52 responses)
  • A
    4% (2)
  • B
    77% (40)
  • C
    8% (4)
  • D
    12% (6)

Why each option

Before applying PCA to a dataset with 150 correlated features having different ranges, it is crucial to scale the data to ensure PCA's effectiveness and minimize impact on model performance.

AUse the Amazon SageMaker built-in algorithm for PCA on the dataset to transform the data.

While Amazon SageMaker has a built-in PCA algorithm, applying it directly to features with 'different ranges' without prior scaling can lead to features with larger ranges dominating the principal components, thereby impacting PCA's effectiveness.

BLoad the data into Amazon SageMaker Data Wrangler. Scale the data with a Min Max ScalerCorrect

Principal Component Analysis (PCA) is highly sensitive to the scale of features; therefore, scaling the data, such as with a Min-Max Scaler in Amazon SageMaker Data Wrangler, is a critical preprocessing step before applying PCA to features with different ranges.

CReduce the dimensionality of the dataset by removing the features that have the highest

Removing features based on 'highest' (e.g., correlation) is a feature selection method, not PCA, and does not address the impact of different feature ranges on PCA.

DReduce the dimensionality of the dataset by removing the features that have the lowest

Removing features based on 'lowest' (e.g., variance) is a feature selection method, not PCA, and does not address the necessary scaling for PCA when features have different ranges.

Concept tested: PCA preprocessing - feature scaling

Source: https://docs.aws.amazon.com/sagemaker/latest/dg/data-wrangler-transformations.html

Topics

#Data Preprocessing#Feature Scaling#Principal Component Analysis (PCA)#Amazon SageMaker Data Wrangler

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full MLS-C01 Practice