MLS-C01 Practice Test 33 — 15 Questions

Question 1

A data engineer is running an Amazon SageMaker Data Wrangler flow on a dataset with 5 million rows. The flow includes several transformations. The engineer wants to validate the data quality by checking for missing values and outliers before training. Which approach is most efficient?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Use Data Wrangler's data quality and insights report to generate a report with statistics and visualizations.. SageMaker Data Wrangler includes a built-in Data Quality and Insights Report that automatically analyzes datasets for missing values, outliers, duplicate rows, and target leakage, producing statistics and visualizations. For a 5-million-row dataset already in a Data Wrangler flow, this is the most efficient native approach because it runs within the same tool without exporting data or spinning up additional compute.

Answer

Export the transformed data to S3 and query with Amazon Athena.

Answer

Use Amazon EMR with Spark to compute statistics.

Answer

Import the data into Amazon QuickSight and create dashboards.

Question 2

A data scientist is training a k-means clustering model on a dataset with 1,000 points. The scientist uses the elbow method to choose the number of clusters. The elbow plot shows a clear bend at k=4. After running k-means with k=4, the scientist wants to evaluate the quality of the clustering. Which THREE of the following are suitable internal clustering validation metrics? (Choose THREE.)

Accepted Answer

Calinski-Harabasz index. The Calinski-Harabasz index (option C) is a suitable internal metric because it evaluates cluster quality using only the data and the resulting labels, computing the ratio of between-cluster dispersion to within-cluster dispersion, so it works without ground-truth labels. The silhouette score (option D) is also internal: it measures how similar each point is to its own cluster versus the nearest other cluster using distances, requiring no external labels. The Davies-Bouldin index (option E) is internal as well, averaging each cluster's worst-case similarity to other clusters based on scatter and separation, with lower values indicating better clustering. Options A and B are not suitable here because the Adjusted Rand index and Rand index are external validation metrics that compare the clustering against a known ground-truth labeling, which the scientist does not have in this unsupervised scenario.

Answer

Adjusted Rand index

Answer

Rand index

Question 3

A data science team at a financial services company is building a fraud detection model using a dataset of credit card transactions. The dataset contains 10 million rows and 20 features, including transaction amount, merchant category, time since last transaction, and customer ID. The target variable 'is_fraud' is highly imbalanced: only 0.1% of transactions are fraudulent. The team is performing exploratory data analysis (EDA) on a sample of 100,000 rows. They compute the correlation matrix and find that 'transaction amount' has a correlation of 0.02 with 'is_fraud'. They also plot the distribution of 'transaction amount' and see that it is heavily right-skewed with a long tail. The team wants to understand the relationship between 'transaction amount' and fraud more deeply before feature engineering. They have access to AWS SageMaker and can run processing jobs. Which course of action is most appropriate?

Accepted Answer

Create bins for 'transaction amount' (e.g., 0-10, 10-50, 50-100, 100+) and compute the fraud rate per bin to detect any non-linear patterns. Creating bins for 'transaction amount' and computing the fraud rate per bin is the most appropriate action because it can reveal non-linear relationships that correlation might miss. Given the heavy right-skew and the near-zero correlation, binning allows the team to see if fraud rates vary across different ranges of transaction amounts, which could inform feature engineering.

Answer

Conclude that 'transaction amount' is not predictive because the correlation is near zero

Answer

Train a random forest model on the sample and use feature importance to assess the predictive power of 'transaction amount'

Answer

Apply a log transformation to 'transaction amount' to reduce skewness and re-run the correlation analysis

Question 4

A company is building a recommendation system using collaborative filtering. The dataset contains implicit feedback (clicks) from users on items. Which algorithm is best suited for this scenario?

Accepted Answer

Alternating Least Squares (ALS). Alternating Least Squares (ALS) is specifically designed for collaborative filtering with implicit feedback data, such as clicks. It handles the implicit nature by treating observed interactions as confidence weights and optimizes the latent factors efficiently. ALS is widely used in recommendation systems for its scalability and ability to handle large sparse matrices.

Answer

Linear Regression

Answer

K-means clustering

Answer

Singular Value Decomposition (SVD)

Question 5

Which TWO metrics are appropriate for evaluating a binary classification model trained on imbalanced data? (Select TWO.)

Accepted Answer

F1 score. The F1 score (B) is appropriate because it is the harmonic mean of precision and recall, so it reflects performance on the minority class rather than being dominated by the majority class as accuracy would be. The precision-recall curve (D) is appropriate because it plots precision against recall across thresholds and is far more informative than ROC curves when the positive class is rare, since it does not include the large number of true negatives. Log loss (A) measures probabilistic calibration and is not a threshold-based classification metric tailored to class imbalance. Accuracy (C) is misleading on imbalanced data because a trivial majority-class predictor can score very high. ROC-AUC (E) can look optimistically high under severe imbalance because the false positive rate is diluted by the large number of negatives, making it less suitable than precision-recall analysis.

Answer

Log loss

Answer

Accuracy

Answer

ROC-AUC

Question 6

Which THREE factors should a data engineer consider when choosing between Amazon S3 and Amazon Redshift for storing large datasets used for machine learning? (Choose 3.)

Accepted Answer

Query performance and latency requirements. Option A is correct because query performance and latency requirements are a primary differentiator: Redshift is a columnar MPP data warehouse designed for low-latency, complex SQL analytics, whereas S3 is object storage where queries run through engines like Athena or Redshift Spectrum and typically incur higher latency, so the workload's latency needs drive the choice. Option C is correct because the cost model differs fundamentally: S3 charges per GB stored with no compute cost, while Redshift charges for provisioned or serverless compute clusters in addition to storage, so a data engineer must weigh storage-versus-compute economics for large ML datasets. Option D is correct because S3 natively stores any format (CSV, Parquet, ORC, JSON, images) with flexible compression, while Redshift is optimized for structured, columnar formats and requires COPY loading, making data format and compression a key selection factor. Option B is not a distinguishing factor because both S3 and Redshift support encryption at rest (SSE-S3, SSE-KMS, and KMS-backed encryption respectively), so it does not help choose between them. Option E is also not a differentiator because both services support lifecycle and retention controls (S3 Lifecycle policies and Redshift snapshots/retention settings), so retention policy alone does not determine the right storage choice.

Answer

Encryption at rest capabilities

Answer

Data retention policies

Question 7

Which TWO AWS services can be used to visualize data distributions as part of exploratory data analysis? (Select TWO.)

Accepted Answer

Amazon QuickSight. Amazon QuickSight is a cloud-native business intelligence service that can visualize data distributions through histograms, box plots, scatter plots, and other chart types, making it suitable for exploratory data analysis. Amazon SageMaker Data Wrangler provides a visual interface to create data distribution charts (e.g., histograms, bar charts) directly within the data preparation workflow, enabling quick inspection of feature distributions before model building.

Answer

AWS Glue

Answer

Amazon Athena

Answer

Amazon Comprehend

Question 8

A data scientist is building a regression model to predict house prices. The dataset contains many features, some of which are highly correlated. The model is overfitting. Which regularization technique should the scientist use to penalize large coefficients and perform feature selection?

Accepted Answer

L1 regularization (Lasso). L1 regularization (Lasso) adds a penalty equal to the absolute value of the coefficients, which can shrink some coefficients to zero, performing feature selection. L2 regularization (Ridge) penalizes squared coefficients but does not zero them out. Elastic Net combines both. Dropout is for neural networks. Option A (L2 regularization) does not perform feature selection. Option B (L1 regularization) is correct for feature selection. Option C (Elastic Net) combines both but L1 alone is simpler for feature selection. Option D (Dropout) is not applicable to linear regression.

Answer

L2 regularization (Ridge)

Answer

Elastic Net regularization

Answer

Dropout

Question 9

A data scientist wants to deploy a PyTorch model for real-time inference with low latency. Which AWS service should they use?

Accepted Answer

Amazon SageMaker real-time endpoint. Amazon SageMaker real-time endpoints are specifically designed for low-latency inference on deployed models, including PyTorch models. They provide persistent HTTPS endpoints that autoscale and support custom containers, making them ideal for real-time prediction workloads.

Answer

Amazon Elastic Container Service (ECS)

Answer

Amazon SageMaker batch transform

Answer

AWS Lambda

Question 10

A company is deploying a machine learning model on SageMaker for real-time inference. The model requires GPU for low latency. Which THREE steps are necessary to set up the endpoint?

Accepted Answer

Create a SageMaker model object that points to the S3 bucket containing the model artifacts and the inference container image. Option C is correct because deploying a real-time endpoint requires a SageMaker model object that references the trained model artifacts in S3 and the inference container image (the image URI and model data URL), which is the deployable entity. Option D is correct because the endpoint configuration defines the production variant, including the GPU instance type such as ml.p3.2xlarge and the initial instance count, which is required to satisfy the low-latency GPU inference requirement. Option E is correct because the endpoint itself must be created from that endpoint configuration to provision the hosting instances and expose the real-time HTTPS inference endpoint. Option A is not necessary for setting up the endpoint itself, since training is a separate lifecycle step and the question assumes a model already exists. Option B is incorrect because batch transform is for offline, asynchronous batch inference and does not create a real-time endpoint.

Answer

Train the model using a SageMaker training job

Answer

Create a SageMaker batch transform job

Question 11

A data scientist is performing exploratory data analysis on a dataset with mixed data types: numerical, categorical, and text. They want to use Amazon SageMaker Data Wrangler to create a quick visualization dashboard. Which set of transformations should they apply in Data Wrangler to handle all data types appropriately?

Accepted Answer

Use the built-in analysis: summary statistics for numerical, word cloud for text, and frequency for categorical.. Amazon SageMaker Data Wrangler provides built-in visualizations tailored to data types: summary statistics for numerical, word cloud for text, and frequency/category plots for categorical. This approach handles mixed data types appropriately without forcing inappropriate transformations. It allows quick exploratory analysis before modeling.

Answer

Convert all features to numerical using one-hot encoding and then create a scatter matrix.

Answer

Apply TF-IDF vectorization to text and then run k-means clustering.

Answer

Use PCA to reduce dimensionality and then visualize the first two components.

Question 12

A data scientist is using Amazon SageMaker to train a deep learning model for natural language processing. The training job is taking too long to converge. The data scientist wants to speed up training without significantly sacrificing model accuracy. Which THREE strategies should the data scientist consider? (Choose three.)

Accepted Answer

Reduce the model size by using fewer layers or smaller hidden dimensions.. Option A is correct because reducing the model size (fewer layers or smaller hidden dimensions) lowers the number of parameters and FLOPs per forward/backward pass, directly decreasing per-epoch training time while typically causing only a modest accuracy trade-off. Option D is correct because mixed precision training with FP16 halves memory bandwidth and storage for activations and weights and leverages NVIDIA Tensor Core throughput, accelerating matrix multiplications on supported GPU instances with minimal accuracy loss when loss scaling is used. Option E is correct because SageMaker distributed data parallelism (e.g., the smdistributed dataparallel library) shards each mini-batch across multiple instances and all-reduces gradients, scaling throughput roughly linearly with instance count while preserving convergence behavior. Option B is not appropriate because multiplying the learning rate by 10 risks divergence or oscillation rather than reliable convergence. Option C is not appropriate because simply maximizing batch size can hurt convergence and generalization and may not fit memory, so it is not a guaranteed speedup without further tuning.

Answer

Increase the learning rate by a factor of 10 to accelerate convergence.

Answer

Increase the batch size to its maximum possible value to utilize GPU memory fully.

Question 13

A data scientist needs to run a one-time SQL query on a large dataset in S3 to create a training dataset. The query involves aggregations and joins. Which service is most suitable?

Accepted Answer

Amazon Athena. Amazon Athena is the correct choice because it is a serverless service that allows running SQL queries directly on data stored in S3, ideal for one-time ad-hoc queries with aggregations and joins. Option A (AWS Glue ETL) is designed for scheduled ETL jobs, not ad-hoc queries. Option C (Amazon EMR with Spark SQL) provides powerful processing but is overkill and requires cluster management. Option D (Amazon RDS) would require moving data into a database, which is inefficient.

Answer

AWS Glue ETL job

Answer

Amazon EMR with Spark SQL

Answer

Amazon RDS with data loaded into it

Question 14

A company is building a machine learning pipeline on AWS. The pipeline includes data ingestion, preprocessing, training, and deployment. Which THREE AWS services can be used to orchestrate the pipeline? (Choose THREE.)

Accepted Answer

AWS Glue Workflows. AWS Glue Workflows (A) is correct because it is a purpose-built orchestration feature of AWS Glue that lets you define and trigger a directed acyclic graph (DAG) of crawlers and ETL jobs, making it suitable for orchestrating the data ingestion and preprocessing stages of an ML pipeline. Amazon SageMaker Pipelines (C) is correct because it is a native CI/CD orchestration service for ML that models each step (processing, training, evaluation, conditional/deployment) as a pipeline step with dependencies, directly covering the full ingestion-to-deployment lifecycle. AWS Step Functions (E) is correct because its state machine language can coordinate Lambda functions, SageMaker jobs, Glue jobs, and other services in sequence, parallel, or with branching/retries, providing general-purpose orchestration across the whole pipeline. Amazon CloudWatch (B) is a monitoring and observability service for metrics, logs, and alarms, not a workflow orchestrator, and AWS Lambda (D) is a compute service for running individual functions that can be invoked by an orchestrator but does not itself provide pipeline orchestration.

Answer

Amazon CloudWatch

Answer

AWS Lambda

Question 15

A machine learning team needs to process a large dataset stored in Amazon S3 using Apache Spark. They want to minimize cost and avoid managing infrastructure. Which AWS service should they use?

Accepted Answer

AWS Glue. AWS Glue is a fully managed, serverless Spark environment that eliminates infrastructure management. It directly meets the requirement to process large datasets in S3 with Apache Spark while minimizing cost, as you only pay for resources consumed during job execution. Glue also integrates natively with S3 and the broader AWS ecosystem, making it the optimal choice for this use case.

Answer

Amazon Athena

Answer

Amazon EMR

Answer

Amazon SageMaker