Amazon Web ServicesAssociate levelMLA-C01

MLA-C01 exam dumps: 457 free AWS Machine Learning Engineer Associate practice questions

Free MLA-C01 practice questions for the AWS Certified Machine Learning Engineer - Associate exam, with the correct answer and a full explanation for every option. Read the first 10 below, browse all 457 by number, or take a timed practice exam.

Question bank last updated December 2024

Start timed practice exam Browse all 457 questions

Free MLA-C01 practice questions

Questions 1 to 10 of 457

Pick an answer before you open the explanation. Each question also has its own page with a permalink.

MLA-C01 Question 1

Select 2

You are building a machine learning pipeline on AWS to predict customer churn. Your training dataset is stored in an S3 bucket and contains several categorical fields, such as 'Customer Type', 'Region', and 'Subscription Plan'. You notice that some of these categorical fields have a high cardinality (e.g., hundreds of unique values). Which of the following preprocessing steps would be the most appropriate for handling these high-cardinality categorical fields?

A
Use one-hot encoding for all categorical fields, including those with high cardinality
B
Apply feature hashing to reduce the dimensionality of high-cardinality categorical fields
C
Use AWS Glue DataBrew to perform frequency binning on high-cardinality categorical fields
D
Replace high-cardinality categorical fields with their mean target value (target encoding)
E
Remove all high-cardinality categorical fields from the dataset to simplify the model

Show answer and explanation

Correct answers: B, D

Explanation

High-cardinality categorical fields can significantly increase the dimensionality of your dataset, posing challenges for model training and efficiency. Feature hashing effectively reduces the dimensionality without losing much information, while target encoding provides a way to incorporate target-related information into the features. Both techniques are valid choices depending on the specific use case. One-hot encoding, while suitable for low-cardinality fields, is not recommended for high-cardinality fields due to feature explosion.

A. Incorrect.
Using one-hot encoding for high-cardinality fields can lead to an explosion in the number of features, making the model computationally expensive and difficult to train.
B. Correct.
Feature hashing is an effective technique to reduce the dimensionality of high-cardinality categorical fields while retaining their information. It is computationally efficient and reduces memory usage.
C. Incorrect.
AWS Glue DataBrew can help preprocess data, but frequency binning is not specifically designed for handling high-cardinality categorical fields. It is more suitable for numerical data.
D. Correct.
Target encoding (replacing with mean target value) is a valid approach for high-cardinality fields as it reduces dimensionality and retains information relevant to the target variable. However, it must be done carefully to avoid data leakage.
E. Incorrect.
Removing high-cardinality categorical fields may result in the loss of valuable information that could improve the model's predictions. This should only be done as a last resort.

MLA-C01 Question 2

Select 3

You are building a machine learning model to predict customer churn for a subscription-based service. The dataset contains missing values, categorical variables, and features with varying scales. Additionally, the dataset contains a 'customer_id' field which uniquely identifies each customer. As part of the data preparation process, which of the following steps should you perform to ensure the data is ready for training? (Choose TWO)

A
Remove the 'customer_id' field from the dataset before training the model.
B
Scale numerical features to ensure they are on a similar range.
C
Fill missing values in the dataset using a unique identifier for each row.
D
Convert categorical variables into numerical representations (e.g., one-hot encoding or label encoding).
E
Normalize the target variable to fall within the range [0, 1].

Show answer and explanation

Correct answers: A, B, D

Explanation

Preparing data for machine learning involves addressing issues like irrelevant features, feature scaling, handling missing values, and encoding categorical variables. Removing identifiers like 'customer_id' prevents noise, scaling ensures all features contribute equally, and encoding categorical variables makes them suitable for model training. Missteps like inappropriate imputation methods or unnecessary normalization of the target variable should be avoided.

A. Correct.
Correct: The 'customer_id' field is not a feature and does not contribute to the prediction task. Keeping it in the dataset can introduce noise or bias, so it must be removed before training.
B. Correct.
Correct: Scaling numerical features helps algorithms (e.g., gradient-based models) perform better by ensuring all features contribute equally to the model's learning process.
C. Incorrect.
Incorrect: Filling missing values with a unique identifier for each row is not a valid imputation technique. Missing values should be handled using statistical methods (e.g., mean, median), domain knowledge, or other appropriate strategies.
D. Correct.
Correct: Machine learning models require numerical inputs. Categorical variables need to be converted into numerical representations for the model to process them effectively.
E. Incorrect.
Incorrect: Normalizing the target variable depends on the problem and algorithm. It is not a required step for data preparation unless the model or problem dictates it.

MLA-C01 Question 3

Select 2

You are building a machine learning pipeline to predict customer churn using Amazon SageMaker. Your dataset contains missing values in several features, and some features have a high cardinality. Which of the following steps should you take to prepare the data for training? (Select TWO.)

A
Impute missing values using statistical methods such as mean, median, or mode.
B
Remove features with missing values to avoid introducing bias into the model.
C
Use one-hot encoding for categorical features with high cardinality.
D
Use embedding layers to represent high-cardinality categorical features.
E
Normalize numerical features to improve model convergence.

Show answer and explanation

Correct answers: A, D

Explanation

Preparing data for machine learning involves addressing missing values and high-cardinality categorical features efficiently. Imputation ensures that missing values do not reduce the dataset's utility, while embedding layers effectively handle high-cardinality features without inflating dimensionality. These steps are crucial for building an effective ML pipeline.

A. Correct.
Imputing missing values using statistical methods is a common and effective way to handle missing data, ensuring the dataset remains usable for training.
B. Incorrect.
Removing features with missing values is not always ideal, as it can lead to loss of potentially important information. Imputation is typically preferred.
C. Incorrect.
One-hot encoding is not efficient for high-cardinality categorical features, as it can significantly increase the dimensionality of the dataset.
D. Correct.
Using embedding layers is a best practice for handling high-cardinality categorical features, as it reduces dimensionality and captures relationships between categories.
E. Incorrect.
Although normalizing numerical features can improve convergence in certain algorithms, this is unrelated to handling missing values or high-cardinality categorical features in this scenario.

MLA-C01 Question 4

Select 3

A data science team is building a machine learning model using Amazon SageMaker. The dataset resides in an Amazon S3 bucket and contains missing values, categorical variables, and numerical features. The team wants to preprocess the data efficiently and ensure it is ready for training. Which steps should the team take to appropriately prepare the dataset for training?

A
Use Amazon SageMaker Data Wrangler to handle missing values, encode categorical variables, and scale numerical features.
B
Write a custom preprocessing script and use SageMaker Processing Jobs to execute it at scale.
C
Directly train the model on the dataset without handling missing values or encoding categorical variables since SageMaker manages preprocessing automatically.
D
Use the Amazon SageMaker Feature Store to impute missing values and transform categorical variables into numerical representations.
E
Leverage AWS Glue to preprocess the data and output a clean dataset to Amazon S3 for model training.

Show answer and explanation

Correct answers: A, B, E

Explanation

Data preparation is essential for building effective machine learning models in SageMaker. Tools like SageMaker Data Wrangler and AWS Glue simplify preprocessing tasks such as handling missing values, encoding categorical variables, and scaling numerical features. Alternatively, custom preprocessing scripts executed using SageMaker Processing Jobs offer flexibility for complex data preparation pipelines. SageMaker itself does not automatically handle preprocessing, and while the Feature Store is useful for feature management, it is not intended for raw data preprocessing.

A. Correct.
Correct: Amazon SageMaker Data Wrangler provides an intuitive interface and built-in capabilities for handling missing values, encoding categorical variables, and scaling numerical features, making it an effective tool for preprocessing datasets.
B. Correct.
Correct: Writing a custom preprocessing script and using SageMaker Processing Jobs allows for flexibility and the ability to handle complex preprocessing tasks at scale.
C. Incorrect.
Incorrect: SageMaker does not handle preprocessing tasks like dealing with missing values or encoding categorical variables automatically. These steps need to be explicitly implemented before training.
D. Incorrect.
Incorrect: While the Amazon SageMaker Feature Store is useful for storing and serving features, it is not primarily used for preprocessing datasets or transforming raw data.
E. Correct.
Correct: AWS Glue can preprocess data at scale and output a clean dataset to Amazon S3, which can then be used for training in SageMaker.

MLA-C01 Question 5

Select 2

You are tasked with building a machine learning model to predict customer churn using data stored in an Amazon S3 bucket. The dataset contains raw data with missing values, inconsistent column names, and a mix of numerical and categorical features. What steps should you take to properly prepare your data for training in Amazon SageMaker?

A
Use AWS Glue to clean the data by normalizing column names and handling missing values before exporting it to Amazon S3.
B
Directly load the raw data from Amazon S3 into SageMaker and handle missing values within the training script.
C
Use Amazon SageMaker Data Wrangler to visually clean, transform, and engineer features before saving the transformed data back to Amazon S3.
D
Use Amazon Redshift to preprocess the data and export the cleaned dataset to Amazon S3.
E
Ignore missing values and proceed with training, as SageMaker automatically handles missing data during model training.

Show answer and explanation

Correct answers: A, C

Explanation

Preparing data for machine learning involves cleaning, transforming, and engineering features to ensure the dataset is suitable for training. AWS Glue and Amazon SageMaker Data Wrangler are purpose-built tools for these tasks. AWS Glue is ideal for large-scale data cleaning and transformation, while SageMaker Data Wrangler provides a visual interface for more interactive feature engineering and preprocessing. Ignoring missing values or relying solely on the training script for preprocessing is not recommended, as it can lead to inefficiencies or suboptimal model performance.

A. Correct.
This is correct. AWS Glue is commonly used to clean and transform raw data at scale, making it an appropriate choice for preparing the dataset.
B. Incorrect.
This is incorrect. While it is possible to handle missing values in the training script, it is not an efficient approach for data preparation, especially if the data requires extensive preprocessing.
C. Correct.
This is correct. Amazon SageMaker Data Wrangler provides a user-friendly interface for cleaning and transforming data, making it a suitable tool for this scenario.
D. Incorrect.
This is incorrect. While Amazon Redshift can be used for data processing, it is primarily a data warehousing solution and not typically the first choice for ML-specific data preparation tasks.
E. Incorrect.
This is incorrect. Ignoring missing data can lead to poor model performance, and SageMaker does not automatically handle missing values unless explicitly configured.

MLA-C01 Question 6

Select 3

You are building a machine learning pipeline to analyze real-time user activity data from a mobile application. The application generates high-velocity JSON logs, which need to be ingested, transformed, and stored in a cost-efficient manner for both batch and real-time analytics. Which combination of services should you use to meet the requirements?

A
Amazon Kinesis Data Streams for ingesting the real-time JSON logs
B
Amazon S3 for storing the transformed logs in a cost-efficient manner
C
Amazon Redshift for processing and storing the raw JSON logs
D
AWS Glue for transforming the JSON logs before storing them
E
Amazon RDS for storing the high-velocity logs in real time

Show answer and explanation

Correct answers: A, B, D

Explanation

To meet the requirements of ingesting, transforming, and storing high-velocity real-time JSON logs, a combination of Amazon Kinesis Data Streams, AWS Glue, and Amazon S3 is ideal. Kinesis handles real-time ingestion, Glue performs transformations, and S3 provides cost-effective and durable storage. Redshift and RDS are not suitable for this use case due to their limitations with high-velocity data ingestion and transformation in real time.

A. Correct.
Correct: Amazon Kinesis Data Streams is designed for ingesting real-time, high-velocity data streams like JSON logs. It provides a scalable solution to process streaming data.
B. Correct.
Correct: Amazon S3 is a cost-efficient and durable storage solution. It is ideal for storing transformed data for batch analytics or archival purposes.
C. Incorrect.
Incorrect: Amazon Redshift is a data warehouse optimized for complex queries and analytics, but it is not suited for processing and directly storing high-velocity raw JSON logs.
D. Correct.
Correct: AWS Glue provides ETL capabilities, making it a good choice for transforming the JSON logs before storing them in the desired format.
E. Incorrect.
Incorrect: Amazon RDS is a relational database service and is not optimized for handling high-velocity real-time data ingestion like JSON logs from a mobile application.

MLA-C01 Question 7

Select 3

You are a Machine Learning Engineer tasked with building a recommendation system for an e-commerce platform. The platform generates large amounts of clickstream data, transaction logs, and user profile data daily. You need to design a data ingestion and storage solution that is cost-effective, scalable, and allows for querying the data in its raw format as well as performing transformations for machine learning model training. Which combination of services should you use?

A
Use Amazon Kinesis Data Streams for real-time ingestion of clickstream data and store it in Amazon S3.
B
Use Amazon Redshift to ingest and store all data, including clickstream, transaction logs, and user profile data.
C
Use AWS Glue to catalog and transform data stored in Amazon S3.
D
Use Amazon S3 as the primary storage for ingested data and configure lifecycle policies for cost optimization.
E
Use Amazon RDS for storing clickstream data and performing complex queries directly on the database.

Show answer and explanation

Correct answers: A, C, D

Explanation

The correct combination of services leverages Amazon S3 as the primary storage solution due to its scalability and cost-effectiveness. Using Amazon Kinesis Data Streams ensures efficient ingestion of real-time data such as clickstream logs, and AWS Glue enables data cataloging and transformation for machine learning purposes. Amazon Redshift and Amazon RDS are not suitable choices for this scenario as they are not optimized for raw data storage or unstructured data handling at scale.

A. Correct.
Amazon Kinesis Data Streams is a suitable choice for real-time ingestion of clickstream data, and Amazon S3 is a cost-effective and scalable storage solution.
B. Incorrect.
Amazon Redshift is a data warehouse solution that is not cost-effective or designed for storing raw data at scale, such as clickstream logs or unstructured data.
C. Correct.
AWS Glue is a serverless data integration service that can catalog and perform ETL on data stored in Amazon S3, making it a good choice for preparing data for machine learning.
D. Correct.
Amazon S3 is a highly scalable and cost-effective storage service. Configuring lifecycle policies allows you to optimize costs by transitioning data to cheaper storage tiers.
E. Incorrect.
Amazon RDS is a relational database service and is not designed for storing large-scale clickstream data or handling unstructured data efficiently.

MLA-C01 Question 8

Select 2

You are working on a machine learning project for a retail company that processes large volumes of transactional data in near real-time. The data needs to be ingested and stored in a way that supports both streaming and batch processing for training your machine learning models. Which of the following services or configurations would best meet these requirements?

A
Use Amazon Kinesis Data Streams to ingest streaming data and store it in Amazon S3 for batch processing.
B
Ingest the data directly into an Amazon RDS database for both real-time and batch processing.
C
Use AWS Glue to continuously transform and load streaming data into Amazon Redshift for analytics queries.
D
Leverage Amazon Managed Streaming for Apache Kafka (Amazon MSK) for real-time ingestion and periodic storage in Amazon S3.
E
Use Amazon Elasticsearch Service to ingest and store the data for both real-time and batch processing.

Show answer and explanation

Correct answers: A, D

Explanation

For a machine learning use case that requires both real-time ingestion and batch processing, using Amazon Kinesis Data Streams or Amazon MSK for ingestion provides robust and scalable solutions. Storing the data in Amazon S3 enables cost-effective batch processing, which is commonly used for training machine learning models. These services are purpose-built for handling high-volume streaming data and complement each other well in this scenario.

A. Correct.
Amazon Kinesis Data Streams is designed for real-time ingestion of streaming data. By storing this data in Amazon S3, you enable cost-effective batch processing for machine learning workflows.
B. Incorrect.
Amazon RDS is not an optimal solution for ingesting high-velocity streaming data. It is better suited for structured data and transactional workloads, not for large-scale real-time ingestion.
C. Incorrect.
AWS Glue is primarily used for ETL processes and does not natively support real-time ingestion of streaming data. It is better suited for batch data transformations.
D. Correct.
Amazon MSK is a managed service for Apache Kafka, which is designed for processing streaming data in real-time. Periodically storing the data in Amazon S3 allows for batch processing, making it an excellent solution for this use case.
E. Incorrect.
Amazon Elasticsearch Service is designed for search and log analytics, not for storing large-scale streaming data or supporting batch processing for machine learning workflows.

MLA-C01 Question 9

Select 2

You are designing a machine learning pipeline for a healthcare company that collects patient data from multiple sources, such as mobile apps, IoT devices, and hospital records. The data is semi-structured and needs to be ingested in real time to support model training and analytics. Which combination of AWS services should you use to ingest, process, and store this data for further analysis and model training?

A
Amazon Kinesis Data Streams for real-time ingestion and Amazon S3 for storage
B
AWS Glue for real-time ingestion and Amazon Redshift for storage
C
Amazon Kinesis Firehose for real-time ingestion and Amazon S3 for storage
D
Amazon DynamoDB Streams for real-time ingestion and Amazon RDS for storage
E
Amazon S3 for real-time ingestion and AWS Lambda for pre-processing

Show answer and explanation

Correct answers: A, C

Explanation

For real-time ingestion of large-scale semi-structured data, Amazon Kinesis Data Streams and Amazon Kinesis Firehose are the most suitable AWS services. Both are designed to handle streaming data and can integrate seamlessly with Amazon S3 for scalable and cost-effective storage. This combination allows you to process and store data efficiently for further analysis and machine learning workflows.

A. Correct.
Correct. Amazon Kinesis Data Streams is designed for real-time ingestion of large-scale data, and Amazon S3 is a cost-effective, scalable storage solution for semi-structured data.
B. Incorrect.
Incorrect. AWS Glue is primarily used for ETL operations and data cataloging, not for real-time ingestion. Amazon Redshift is suitable for analytics but not ideal for semi-structured data storage.
C. Correct.
Correct. Amazon Kinesis Firehose is optimized for real-time data ingestion and can directly deliver data to Amazon S3 for storage.
D. Incorrect.
Incorrect. While DynamoDB Streams supports real-time data processing, it is not suitable for large-scale semi-structured data ingestion. Amazon RDS is a relational database service and not ideal for storing semi-structured data.
E. Incorrect.
Incorrect. Amazon S3 is a storage service and does not support real-time ingestion. AWS Lambda can perform pre-processing but is not a dedicated real-time ingestion service.

MLA-C01 Question 10

Select 2

You are tasked with building a machine learning pipeline to process large volumes of structured and unstructured data from multiple sources, such as IoT devices and transactional databases. The data needs to be ingested in near real-time and stored in a scalable, fault-tolerant manner for downstream ML processing. Which combination of services should you use to meet these requirements?

A
Amazon Kinesis Data Streams for real-time data ingestion and Amazon S3 for scalable storage
B
AWS Lambda for real-time data ingestion and Amazon DynamoDB for storage
C
Amazon Kinesis Firehose for real-time data transformation and Amazon S3 for storage
D
Amazon Redshift for real-time ingestion and storage
E
AWS Glue for real-time ingestion and Amazon S3 for storage

Show answer and explanation

Correct answers: A, C

Explanation

For this use case, you need real-time data ingestion and scalable storage to handle both structured and unstructured data from multiple sources. Amazon Kinesis Data Streams and Amazon Kinesis Firehose are both designed for real-time ingestion, while Amazon S3 provides the scalability and durability needed for storage. AWS Lambda, Amazon DynamoDB, Amazon Redshift, and AWS Glue do not fully meet the requirements of high-throughput, real-time ingestion or scalable storage for large volumes of unstructured data.

A. Correct.
Amazon Kinesis Data Streams is designed for real-time ingestion of high-throughput data, and Amazon S3 provides scalable, durable storage that is ideal for storing large volumes of structured and unstructured data for ML workloads.
B. Incorrect.
AWS Lambda is a serverless compute service and is not primarily used for high-throughput real-time data ingestion. While Amazon DynamoDB is a scalable NoSQL database, it is not the best choice for storing large volumes of unstructured data.
C. Correct.
Amazon Kinesis Firehose is a managed service that can perform real-time data transformation and directly deliver the data to Amazon S3, making it an ideal choice for this use case.
D. Incorrect.
Amazon Redshift is a data warehousing service that is optimized for analytics, not for real-time ingestion or storing unstructured data.
E. Incorrect.
AWS Glue is primarily a data integration service for ETL workflows, not for real-time data ingestion. While Amazon S3 is suitable for storage, Glue does not meet the real-time ingestion requirement.

Timed practice exam

Take a MLA-C01 practice test under exam conditions

65 questions in 130 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam

MLA-C01 practice questions 1 to 100 of 457

Every question has a page with the answer and explanation. Numbers are stable, so you can bookmark or share them. The bank is split into 5 pages of up to 100 questions.

MLA-C01 exam dumps FAQ

Are these MLA-C01 dumps real exam questions?

No. These are original practice questions written to the AWS Certified Machine Learning Engineer - Associate exam objectives, not questions copied from a live exam. Memorising leaked questions violates Amazon Web Services's candidate agreement and stops working the moment the question pool rotates. Use this bank to check your understanding of each domain and to find the topics you still need to study.

How many MLA-C01 practice questions are there?

457 questions, each with the correct answer, an explanation of the answer, and a note on why every other option is wrong. The first 10 are on this page and every question has its own page linked below.

Are the MLA-C01 exam dumps free?

Yes. Every question, answer and explanation on this page and the linked question pages is free to read without an account. A free HydraNode account adds timed practice exams, scoring and progress tracking across attempts.

How do I take a timed MLA-C01 practice test?

Sign in and start the AWS Certified Machine Learning Engineer - Associate exam on HydraNode. A session gives you 65 questions drawn from this bank in 130 minutes, then a score report with a per-question review.