MLA-C01 Data Preparation for Machine Learning — Timed 10 Questions

Question 1

A company is using AWS Glue to prepare data for a machine learning pipeline. The source data is in an Amazon S3 bucket in CSV format. The data scientist wants to convert the data to Parquet format and partition it by date. Which AWS Glue feature should be used to optimize the data for query performance and reduce storage costs?

10 minute time limit — choose an answer to begin.

Accepted Answer

Use AWS Glue DynamicFrame to repartition the data and write it as Parquet.. AWS Glue DynamicFrames provide built-in optimizations for writing data in columnar formats like Parquet, which improves query performance through predicate pushdown and compression, and reduces storage costs by using efficient encoding. The DynamicFrame's `repartition()` method allows you to control the number of output files, and writing as Parquet directly from Glue avoids intermediate conversions, making it the most efficient choice for this task.

Answer

Use Amazon Athena to convert the data to JSON format and store it in S3.

Answer

Use AWS Glue to convert the data to Apache Hive format.

Answer

Use Apache Spark DataFrame to write the data as CSV with Snappy compression.

Question 2

A machine learning engineer is preparing a dataset for a binary classification model. The dataset has a severe class imbalance (95% class A, 5% class B). The engineer wants to use Amazon SageMaker to train the model. Which data preparation technique should the engineer apply to the training dataset to address the imbalance and improve model performance?

Accepted Answer

Apply Synthetic Minority Over-sampling Technique (SMOTE) to generate synthetic samples for the minority class.. SMOTE generates synthetic samples for the minority class by interpolating between existing minority instances, which directly addresses the severe class imbalance (95% class A, 5% class B) by creating a more balanced training dataset. This technique is particularly effective for tabular data in Amazon SageMaker, as it increases the representation of the minority class without simply duplicating existing samples, thereby reducing overfitting and improving the model's ability to learn decision boundaries for the minority class.

Answer

Apply data augmentation to the majority class by adding noise.

Answer

Use a weighted loss function during training to penalize misclassifications of the minority class.

Answer

Apply random under-sampling to reduce the majority class to match the minority class size.

Question 3

A company uses AWS Glue to run ETL jobs that prepare data for machine learning. The data is stored in Amazon S3 in Parquet format. A data engineer notices that the Glue job is running slowly and consuming a lot of resources. What is the MOST cost-effective way to improve the performance of the Glue job?

Accepted Answer

Use the G.1X worker type, which provides more memory per worker compared to the Standard worker type.. Increasing the number of DPUs (Data Processing Units) in AWS Glue can improve parallelism and reduce job runtime, but it increases cost. Using G.1X worker type with more memory per worker can improve performance without increasing DPU count, offering better resource utilization. Switching to CSV may degrade performance. Using partition pruning on the source data can reduce data scanned but may not address resource consumption.

Answer

Use partition pruning on the source data to reduce the amount of data processed.

Answer

Switch the output format from Parquet to CSV to reduce processing overhead.

Answer

Use a larger instance type for the Glue job by increasing the number of DPUs.

Question 4

A machine learning team is building a model using a dataset that contains a mix of numerical and categorical features. The categorical features have high cardinality (e.g., zip code with thousands of unique values). The team wants to use Amazon SageMaker for training. Which technique should the team use to encode the high-cardinality categorical features effectively?

Accepted Answer

Apply target encoding (mean encoding) to the high-cardinality features.. Target encoding replaces each high-cardinality category with a statistic (typically the mean of the target) computed from the training data, which compresses thousands of zip codes into a single informative numeric feature without exploding dimensionality. It is the standard SageMaker-recommended approach for high-cardinality categoricals because it preserves predictive signal while keeping the feature space small.

Answer

Apply hash encoding to map categories to a fixed number of buckets.

Answer

Apply one-hot encoding to all categorical features.

Answer

Apply label encoding to assign integer values to each category.

Question 5

A machine learning team is preparing a dataset for a regression model. The dataset contains numerical features that are on different scales (e.g., age 0-100, income 0-1,000,000). The team plans to use Amazon SageMaker to train a linear regression model. Which THREE data preparation steps should the team take to ensure the model performs well? (Select THREE.)

Accepted Answer

Handle missing values by imputation or removal.. Option C is correct because missing values in a dataset will cause errors or bias in SageMaker linear regression training, so imputing them (e.g., mean/median) or removing affected rows ensures a complete numeric matrix for the algorithm. Option D is correct because linear regression in SageMaker requires all input features to be numeric, so any categorical variables must be converted via one-hot encoding (or similar numeric encoding) before training. Option E is correct because the scenario explicitly states features are on very different scales (age 0-100 vs. income 0-1,000,000); standardization (z-score) or min-max normalization puts them on comparable scales, which improves convergence and coefficient stability for linear regression. Option A is not required: feature selection can help reduce overfitting or dimensionality, but it is not a mandatory preparation step to make linear regression perform well with differently scaled numeric features. Option B is not required: outlier removal can be beneficial in some cases, but it is not a necessary data preparation step for the stated goal and may discard valid data; the question asks for steps the team should take to ensure good performance, and scaling, missing-value handling, and numeric encoding are the essential ones.

Answer

Apply feature selection to reduce the number of features.

Answer

Remove outliers from the dataset.

Question 6

A data scientist is using Amazon SageMaker Data Wrangler to create a data flow for a machine learning project. The source data is in Amazon S3 and contains PII (personally identifiable information) such as email addresses and credit card numbers. The data scientist needs to prepare the data for training while ensuring compliance with data privacy regulations. Which THREE actions should the data scientist take? (Select THREE.)

Accepted Answer

Use Data Wrangler to redact or remove PII columns from the dataset before training.. Option B is correct because SageMaker Data Wrangler provides built-in transforms (such as 'Drop column' or redaction/obfuscation transforms) that let the data scientist remove or mask PII columns like email addresses and credit card numbers before the data reaches training, directly satisfying data minimization requirements under privacy regulations. Option E is correct because Data Wrangler supports anonymization transforms such as hashing (e.g., SHA-256 via the 'Encode categorical' or custom Pandas/PySpark transform) to irreversibly pseudonymize PII while preserving the column for model utility, which is a standard privacy-preserving technique. Option D is correct because writing the prepared output to an S3 bucket with server-side encryption (SSE-S3, SSE-KMS, or SSE-C) protects the data at rest and is a required control for handling sensitive data under regulations such as GDPR, HIPAA, or PCI DSS. Option A is incorrect because relying on the model not to memorize PII is not a compliant control—models can and do memorize training data, and raw PII in training sets violates data minimization and privacy requirements. Option C is incorrect because copying data to another bucket without any transformation neither removes nor protects the PII, so it does nothing to achieve compliance.

Answer

Include the raw PII in the training dataset and rely on the model to not memorize it.

Answer

Use AWS Glue to copy the data to a separate bucket without any transformations.

Question 7

A data scientist is preparing a large dataset for training a binary classification model. The dataset has a severe class imbalance (95% negative, 5% positive). Which data preparation technique should the scientist use to address this imbalance without losing too much data?

Accepted Answer

SMOTE (Synthetic Minority Over-sampling Technique). SMOTE (Synthetic Minority Over-sampling Technique) is the best choice because it generates synthetic examples for the minority class by interpolating between existing minority instances and their k-nearest neighbors, rather than simply duplicating data. This addresses the severe 95:5 class imbalance without losing data (as undersampling would) and without the overfitting risk of naive random oversampling. The synthetic samples help the model learn a more general decision boundary for the positive class.

Answer

Random undersampling of the majority class

Answer

Random oversampling of the minority class

Answer

Apply class weights during model training

Question 8

A machine learning engineer is preparing a dataset that contains both numerical and categorical features. The categorical features have high cardinality (e.g., zip code with thousands of unique values). Which technique is most appropriate for encoding these high-cardinality categorical features?

Accepted Answer

Target encoding. Target encoding is the most appropriate technique for high-cardinality categorical features because it replaces each category with the mean of the target variable for that category, effectively capturing the predictive signal while keeping the feature as a single numeric column. This avoids the dimensionality explosion of one-hot encoding and the arbitrary ordinality of label encoding, making it a common choice in gradient boosting frameworks like XGBoost or LightGBM for datasets with thousands of unique categories.

Answer

Label encoding

Answer

One-hot encoding

Answer

Frequency encoding

Question 9

A team is building a regression model on a dataset with missing values in multiple features. They decide to use a k-Nearest Neighbors (k-NN) imputer. The dataset has 100,000 rows and 50 features. Which step should the team take to ensure the imputation is efficient and accurate?

Accepted Answer

Standardize the features before applying k-NN imputation. Standardizing features before applying k-NN imputation is critical because k-NN relies on distance calculations (e.g., Euclidean distance). If features are on different scales (e.g., one feature ranges 0–1 and another 0–100,000), the distance metric will be dominated by the larger-scale feature, leading to biased neighbor selection and inaccurate imputation. Standardization (e.g., z-score scaling) ensures each feature contributes equally to the distance computation, improving both efficiency and accuracy.

Answer

Set k=1 to minimize bias

Answer

Use all 100,000 rows to find neighbors for each missing value

Answer

Use only the feature with missing values to find neighbors

Question 10

A data engineer is using Amazon SageMaker Data Wrangler to prepare a dataset. The dataset contains a column 'review_date' with timestamps. The engineer wants to extract the day of the week as a new feature. How should this transformation be performed in Data Wrangler?

Accepted Answer

Use the 'day_of_week' transform on the 'review_date' column. Amazon SageMaker Data Wrangler includes a built-in 'day_of_week' transform that directly extracts the day of the week (e.g., Monday, Tuesday) from a timestamp column without requiring custom code or additional formatting. This transform is optimized for Data Wrangler's visual interface and integrates seamlessly with its processing pipeline.

Answer

Write a custom Python script using pandas dt.day_name()

Answer

Use one-hot encoding on the timestamp

Answer

Use the 'extract' transform with format '%A'