Skip to content
BytePatterns

CLF-C02 · Domain 3: Cloud Technology and Services · 34% of the exam

Task 3.7: Identify AWS artificial intelligence and machine learning (AI/ML) services and analytics services.

Which AI/ML service does which job, and the services that query, stream, transform and visualise data.

Study it

  • AI/ML and analytics services

    Lesson coming

Sample questions

Try each one before opening the answer. Every option is explained, with the AWS documentation page that proves it.

Question 1 · choose 1

An analyst wants to run standard SQL queries directly against CSV and Parquet files already stored in Amazon S3. She does not want to load the data into a database or manage any servers, and she wants to pay only for the queries she runs. Which service fits?

  1. AAmazon Kinesis Data Streams
  2. BAmazon Quick Sight
  3. CAmazon Athena
  4. DAmazon SageMaker AI
Show the answer and why

  • AAmazon Kinesis Data Streams

    Incorrect

    Kinesis Data Streams collects and processes streaming data in real time. It does not query files already sitting in S3.

  • BAmazon Quick Sight

    Incorrect

    Quick Sight builds interactive dashboards and visualizations for business intelligence. It is not the SQL query engine over S3 files.

  • CAmazon Athena

    Correct

    Athena is a serverless, interactive query service for analyzing data directly in Amazon S3 with standard SQL; you pay only for the queries you run.

  • DAmazon SageMaker AI

    Incorrect

    SageMaker AI is for building, training and deploying machine learning models, not for ad-hoc SQL over files.

Point a query engine at the files where they are: Athena reads S3 directly, with no cluster to run and no data to load first.

AWS documentation

Task 3.7Report a problem with this question

Question 2 · choose 1

A bank wants to add a chatbot to its website and phone line that understands what customers type or say, asks follow-up questions, and helps them check a balance or report a lost card. Which AWS service is designed to build this conversational interface?

  1. AAmazon Polly
  2. BAmazon Comprehend
  3. CAmazon Rekognition
  4. DAmazon Lex
Show the answer and why

  • AAmazon Polly

    Incorrect

    Polly turns text into lifelike speech. It can give a bot a voice, but it does not understand what the customer means.

  • BAmazon Comprehend

    Incorrect

    Comprehend uses natural language processing to find insights in text, such as sentiment and key phrases. It does not hold a conversation.

  • CAmazon Rekognition

    Incorrect

    Rekognition analyzes images and videos. It does not process typed or spoken conversations.

  • DAmazon Lex

    Correct

    Lex builds conversational interfaces using voice and text: it recognizes what the user wants and manages the dialog to fulfill the request.

Lex is the bot; Polly can be its voice, Comprehend can analyze the transcripts, and Rekognition works on pictures, not words.

AWS documentation

Task 3.7Report a problem with this question

Question 3 · choose 1

A shopping website wants to collect every click from its visitors as a continuous stream and process the records within seconds, for example to update live dashboards. Which service is designed to ingest this data?

  1. AAmazon Athena
  2. BAmazon Quick Sight
  3. CAmazon Kinesis Data Streams
  4. DAmazon Lex
Show the answer and why

  • AAmazon Athena

    Incorrect

    Athena runs SQL queries on data already stored in Amazon S3. It does not take in a live stream of records.

  • BAmazon Quick Sight

    Incorrect

    Quick Sight builds dashboards and visualizations; it needs the data to be collected somewhere first.

  • CAmazon Kinesis Data Streams

    Correct

    Kinesis Data Streams collects and processes large streams of data records, such as web clickstreams, in real time.

  • DAmazon Lex

    Incorrect

    Lex builds conversational chatbots. It is not a data streaming service.

Data that arrives continuously and must be acted on within seconds is a streaming problem; Kinesis Data Streams is the intake.

AWS documentation

Task 3.7Report a problem with this question

Question 4 · choose 1

A data team must discover the data sets spread across its sources, catalog their schemas, and build extract, transform and load (ETL) jobs that clean and combine the data into its data lake — without managing servers. Which service fits?

  1. AAWS Glue
  2. BAmazon Kinesis Data Streams
  3. CAmazon Quick Sight
  4. DAmazon Polly
Show the answer and why

  • AAWS Glue

    Correct

    Glue is a serverless data integration service: it discovers data, keeps a central Data Catalog, and runs ETL pipelines to load data into data lakes.

  • BAmazon Kinesis Data Streams

    Incorrect

    Kinesis Data Streams ingests real-time streams of records. It does not catalog data sets or run ETL jobs over them.

  • CAmazon Quick Sight

    Incorrect

    Quick Sight visualizes data in dashboards. Preparing and cataloging the data comes before it.

  • DAmazon Polly

    Incorrect

    Polly turns text into speech. It is unrelated to data integration.

Discover, catalog, transform and load: Glue. The cataloged data can then be queried with Athena or shown in Quick Sight.

AWS documentation

Task 3.7Report a problem with this question

Question 5 · choose 1

A team of data scientists wants to build, train and deploy its own machine learning models to predict customer churn, using managed infrastructure instead of setting up its own servers. Which service is designed for this?

  1. AAmazon Rekognition
  2. BAmazon Comprehend
  3. CAmazon Polly
  4. DAmazon SageMaker AI
Show the answer and why

  • AAmazon Rekognition

    Incorrect

    Rekognition is a ready-made service for analyzing images and videos; it is not a platform for building your own models.

  • BAmazon Comprehend

    Incorrect

    Comprehend is a ready-made natural language processing service for finding insights in text.

  • CAmazon Polly

    Incorrect

    Polly converts text into lifelike speech.

  • DAmazon SageMaker AI

    Correct

    SageMaker AI is a fully managed service to build, train and deploy machine learning models.

Ready-made AI services solve one task each; when you need your own model, trained on your own data, the platform is SageMaker AI.

AWS documentation

Task 3.7Report a problem with this question

Practise domain 3 →Practise all domains →