Microsoft Azure AI Fundamentals practice questions

1. Question 1 of 25

   Select 2

   In which two scenarios can you use a speech synthesis solution? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

   • Aan automated voice that reads back a credit card number entered into a telephone by using a numeric keypad
   • Bgenerating live captions for a news broadcast
   • Cextracting key phrases from the audio recording of a meeting
   • Dan AI character in a computer game that speaks audibly to a player

   Show answer

   Correct answer: A, D

   How others answered

   • AD100%

   Explanation

   Speech recognition - the ability to detect and interpret spoken input. Speech synthesis - the ability to generate spoken output.

   Adapted from community discussion by Kline · 2024-04-02 · 9 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 1 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. K

      Kline

      2024-04-02

      9 upvotes

      Speech recognition - the ability to detect and interpret spoken input. Speech synthesis - the ability to generate spoken output.

   2. N

      neptune2000

      2022-08-13

      Selected A, D7 upvotes

      A and D is correct --> https://docs.microsoft.com/en-us/azure/cognitive-services/speech-service/text-to-speech

   3. S

      Sylph

      2024-02-19

      2 upvotes

      Text to speech enables your applications, tools, or devices to convert text into humanlike synthesized speech. The text to speech capability is also known as speech synthesis. https://learn.microsoft.com/en-us/azure/ai-services/speech-service/text-to-speech

   4. V

      valv

      2024-10-15

      1 upvote

      The thing that made me uncertain about A is that if the telephone call did NOT have a reasonable expectation of privacy. If it was on a speaker phone, in public, would the bot reading out your CC number out loud be considered as breaking the principle of Security & Privacy for Responsible AI? Maybe I am overthinking it and these exams sometimes really do just want the answer based on the simplest scenario with no extra context.

2. Question 2 of 25

   HOTSPOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Each correct selection is worth one point. Hot Area:

   

   This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

   Show the answer figure

   Answer

   Explanation

   The translator service provides multi-language support for text translation, transliteration, language detection, and dictionaries. Speech-to-Text, also known as automatic speech recognition (ASR), is a feature of Speech Services that provides transcription. Reference: https://docs.microsoft.com/en-us/azure/cognitive-services/Translator/translator-info-overview https://docs.microsoft.com/en-us/legal/cognitive-services/speech-service/speech-to-text/transparency-note

   Adapted from community discussion by idioteque · 2022-08-18 · 11 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • azure.microsoft.com documentation
   • azure.microsoft.com documentation

   Answer figure

   

   View community discussion4

   Candidate context

   Question 2 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. I

      idioteque

      2022-08-18

      11 upvotes

      I just took the exam today 8/19/2022. The main keywords here Translator service are now replaced by "Language Service". So I got confused on the last question and to my understanding based on the exam, the "language service" covers as a whole (NLP, Speech, QnA Maker, Text service) all related past questions from exam topics. So for this particular question. I answered YYY. I still managed to pass the exam with the score of 916/1000.

   2. V

      vcehunger

      2022-02-12

      7 upvotes

      Should be YYY https://azure.microsoft.com/en-us/services/cognitive-services/speech-to-text/#overview

   3. U

      UM68

      2022-02-24

      4 upvotes

      This should be correct, because you require Speech first in addition to convert audio to text, see more details here: https://azure.microsoft.com/en-us/services/cognitive-services/speech-translation/#features

   4. R

      rdemontis

      2023-06-01

      3 upvotes

      Given answer is correct. to convert audible speech in to text you need speech to text service

3. Question 3 of 25

   In which scenario should you use key phrase extraction?

   • Aidentifying whether reviews of a restaurant are positive or negative
   • Bgenerating captions for a video based on the audio track
   • Cidentifying which documents provide information about the same topics
   • Dtranslating a set of documents from English to German

   Show answer

   Correct answer: C

   How others answered

   • C92%

   Explanation

   C - "Use key phrase extraction to quickly identify the main concepts in text. For example, in the text "The food was delicious and the staff were wonderful.", key phrase extraction will return the main topics: "food" and "wonderful staff"."

   Adapted from community discussion by Radoslavov · 2021-11-24 · 11 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 3 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. R

      Radoslavov

      2021-11-24

      11 upvotes

      C - "Use key phrase extraction to quickly identify the main concepts in text. For example, in the text "The food was delicious and the staff were wonderful.", key phrase extraction will return the main topics: "food" and "wonderful staff"."

   2. B

      BenLam

      2024-03-05

      2 upvotes

      Given the example by Microsoft in the key phrase extraction. https://learn.microsoft.com/en-us/azure/ai-services/language-service/key-phrase-extraction/how-to/call-api we assume the answer is A but as others have pointed out it is C as Sentiment Analysis is needed if it is option A. https://learn.microsoft.com/en-us/azure/ai-services/language-service/sentiment-opinion-mining/overview?tabs=prebuilt

   3. G

      globy118

      2023-07-08

      2 upvotes

      A In which scenario should you use a key phrase extraction? Use key phrase extraction to quickly identify the main concepts in text. For example, in the text "The food was delicious and the staff were wonderful.", key phrase extraction will return the main topics: "food" and "wonderful staff" https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/key-phrase-extraction/overview

   4. M

      meena_lucky

      2023-07-09

      3 upvotes

      A says "positive or negative review' which is definitely sentiment analysis and not key phrase extraction

4. Question 4 of 25

   You are building an AI system. Which task should you include to ensure that the service meets the Microsoft transparency principle for responsible AI?

   • AEnsure that all visuals have an associated text that can be read by a screen reader.
   • BEnable autoscaling to ensure that a service scales based on demand.
   • CProvide documentation to help developers debug code.
   • DEnsure that a training dataset is representative of the population.

   Show answer

   Correct answer: C

   How others answered

   • C97%

   Explanation

   C is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#transparency Achieving transparency helps the team to understand the data and algorithms used to train the model, what transformation logic was applied to the data, the final model generated, and its associated assets. This information offers insights about how the model was created, which allows it to be reproduced in a transparent way. Snapshots within Azure Machine Learning workspaces support transparency by recording or retraining all training-related assets and metrics involved in the experiment.

   Adapted from community discussion by zellck · 2023-06-24 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 4 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. D

      dolphin_1923

      2024-09-19

      26 upvotes

      C is correct. https://docs.microsoft.com/en-us/learn/modules/get-started-ai-fundamentals/7-understand-responsible-ai

   2. B

      BenHung

      2024-09-19

      5 upvotes

      The Microsoft transparency principle for responsible AI requires that AI systems are understandable by users and operators. Therefore, the task that directly aligns with this principle is: D. Ensure that a training dataset is representative of the population. This helps to ensure that the AI system’s decisions are fair and unbiased, which is a key aspect of transparency. Option A, while important for accessibility, does not directly relate to the transparency of an AI system. Option B is more about the performance and scalability of the system, and option C, while important for maintainability, does not directly contribute to the transparency of the AI system from the user’s perspective. Therefore, the most appropriate answer is D.

   3. Z

      zellck

      2023-06-24

      Selected C2 upvotes

      C is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#transparency Achieving transparency helps the team to understand the data and algorithms used to train the model, what transformation logic was applied to the data, the final model generated, and its associated assets. This information offers insights about how the model was created, which allows it to be reproduced in a transparent way. Snapshots within Azure Machine Learning workspaces support transparency by recording or retraining all training-related assets and metrics involved in the experiment.

   4. C

      CLVASQUEZ

      2024-04-25

      3 upvotes

      While options A, B, and C are important aspects of building an AI system, they do not directly contribute to the transparency of the AI system as defined by the principle. Option A is more related to accessibility, option B is about scalability, and option C is about providing support for developers, none of which directly ensure transparency in AI decision-making. Therefore, option D is the most appropriate choice to ensure the service meets the Microsoft transparency principle for responsible AI.

5. Question 5 of 25

   You use natural language processing to process text from a Microsoft news story. You receive the output shown in the following exhibit.

   

   Which type of natural languages processing was performed?

   • Aentity recognition
   • Bkey phrase extraction
   • Csentiment analysis
   • Dtranslation

   Show answer

   Correct answer: A

   How others answered

   • A100%

   Explanation

   Entity recognition has category classification, key phrase extraction does not. Take a look at Key Phrase extraction results: [https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/how-tos/text-analytics-how-to-keyword-extraction] with NER results: [https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/how-tos/text-analytics-how-to-entity-linking?tabs=version-3-preview]

   Adapted from community discussion by digitallycan · 2021-11-13 · 24 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn
   • azure.microsoft.com documentation
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 5 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. D

      digitallycan

      2021-11-13

      24 upvotes

      Entity recognition has category classification, key phrase extraction does not. Take a look at Key Phrase extraction results: [https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/how-tos/text-analytics-how-to-keyword-extraction] with NER results: [https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/how-tos/text-analytics-how-to-entity-linking?tabs=version-3-preview]

   2. K

      kty

      2022-03-26

      6 upvotes

      I agree Extract a broad range of pre-built entities such as people, places, organizations, date/time, numerals and over 100 types of personally identifiable information (PII), including protected health information (PHI), in documents using named entity recognition. Quickly evaluate and identify the main points in unstructured text. Get a list of relevant phrases that best describe the subject of each record using key phrase extraction. Easily organize information to make sense of important topics and trends.

   3. I

      imcee_

      2021-10-16

      6 upvotes

      it is entity recognition https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/named-entity-types?tabs=general

   4. E

      ExamTopics007

      2021-12-04

      2 upvotes

      Entity Recognition. You can provide the Text Analytics service with unstructured text and it will return a list of entities in the text that it recognizes. You can provide the Text Analytics service with unstructured text and it will return a list of entities in the text that it recognizes. The service can also provide links to more information about that entity on the web. An entity is essentially an item of a particular type or a category; and in some cases, subtype, such as those as shown in the following table. https://docs.microsoft.com/en-us/learn/modules/analyze-text-with-text-analytics-service/2-get-started-azure

6. Question 6 of 25

   You have a natural language processing (NLP) model that was created by using data obtained without permission. Which Microsoft principle for responsible AI does this breach?

   • Areliability and safety
   • Bprivacy and security
   • Cinclusiveness
   • Dtransparency

   Show answer

   Correct answer: B

   How others answered

   • B100%

   Explanation

   In this case, creating an NLP model using data obtained without permission breaches the principle of **privacy and security**. Source: Conversation with Bing, 4/26/2023 (1) Responsible and trusted AI - Cloud Adoption Framework | Microsoft Learn. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai Accessed 4/26/2023. (2) Responsible AI principles from Microsoft. https://www.microsoft.com/en-us/ai/responsible-ai Accessed 4/26/2023. (3) Responsible AI principles from Microsoft. https://www.microsoft.com/en-gb/ai/responsible-ai Accessed 4/26/2023. (4) Identify guiding principles for responsible AI - Training. https://learn.microsoft.com/en-us/training/modules/responsible-ai-principles/ Accessed 4/26/2023. (5) Responsible AI - Cloud Adoption Framework | Microsoft Learn. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/strategy/responsible-ai Accessed 4/26/2023.

   Adapted from community discussion by master_yoda · 2023-04-26 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 6 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. M

      master_yoda

      2023-04-26

      Selected B4 upvotes

      In this case, creating an NLP model using data obtained without permission breaches the principle of **privacy and security**. Source: Conversation with Bing, 4/26/2023 (1) Responsible and trusted AI - Cloud Adoption Framework | Microsoft Learn. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai Accessed 4/26/2023. (2) Responsible AI principles from Microsoft. https://www.microsoft.com/en-us/ai/responsible-ai Accessed 4/26/2023. (3) Responsible AI principles from Microsoft. https://www.microsoft.com/en-gb/ai/responsible-ai Accessed 4/26/2023. (4) Identify guiding principles for responsible AI - Training. https://learn.microsoft.com/en-us/training/modules/responsible-ai-principles/ Accessed 4/26/2023. (5) Responsible AI - Cloud Adoption Framework | Microsoft Learn. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/strategy/responsible-ai Accessed 4/26/2023.

   2. X

      XtraWest

      2023-04-07

      Selected B7 upvotes

      Breach of the Microsoft principle for responsible AI known as Privacy, Security, and Compliance

   3. R

      rdemontis

      2023-06-08

      Selected B4 upvotes

      For me the answer is B because it is related to a privacy concern: using data without permission

   4. C

      Certification_Champs

      2023-10-22

      3 upvotes

      B. privacy and security Using data obtained without permission breaches the Microsoft principle for responsible AI related to "privacy and security." This principle emphasizes the importance of respecting the privacy of individuals and securing their data. Data should be collected and used in a legal and ethical manner, and obtaining data without proper consent or permission can lead to privacy and security violations.

7. Question 7 of 25

   DRAG DROP - Match the services to the appropriate descriptions. To answer, drag the appropriate service from the column on the left to its description on the right. Each service may be used once, more than once, or not at all. You may need to drag the split bar between panes or scroll to view content. NOTE: Each correct selection is worth one point.

   

   This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

   Show the answer figure

   Answer

   Explanation

   1. Language Service 2. Speech https://learn.microsoft.com/en-us/training/modules/build-faq-chatbot-qna-maker-azure-bot-service/2-get-started-knowledge-base The Language service includes a custom question answering feature that enables you to create a knowledge base of question and answer pairs that can be queried using natural language input. https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/overview The Speech service provides speech to text and text to speech capabilities with an Speech resource. You can transcribe speech to text with high accuracy, produce natural-sounding text to speech voices, translate spoken audio, and use speaker recognition during conversations.

   Adapted from community discussion by zellck · 2023-12-24 · 5 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn

   Answer figure

   

   View community discussion4

   Candidate context

   Question 7 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. Z

      zellck

      2023-12-24

      5 upvotes

      1. Language Service 2. Speech https://learn.microsoft.com/en-us/training/modules/build-faq-chatbot-qna-maker-azure-bot-service/2-get-started-knowledge-base The Language service includes a custom question answering feature that enables you to create a knowledge base of question and answer pairs that can be queried using natural language input. https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/overview The Speech service provides speech to text and text to speech capabilities with an Speech resource. You can transcribe speech to text with high accuracy, produce natural-sounding text to speech voices, translate spoken audio, and use speaker recognition during conversations.

   2. C

      ColoradoChick

      2024-01-20

      6 upvotes

      QnA Maker (now Language) = creates the Knowledge Base Bot Service - uses the Knowledge Base than Language created The question asks "... to query a knowledge base"

   3. R

      rdemontis

      2023-12-08

      4 upvotes

      Correct answer should be Language Service. The Language Service enables the use of natural language to query a knowledge base. It provides capabilities for natural language understanding, including processing and analyzing text to extract meaning, detect intent, and perform various language-related tasks such as sentiment analysis, language translation, and text classification. It allows developers to build applications that can understand and respond to user queries expressed in natural language, making it suitable for querying knowledge bases and extracting relevant information. (Chat GPT)

   4. R

      rdemontis

      2023-12-14

      2 upvotes

      I reevaluate my previous answer. I honestly feel that the given answer is correct. At the end of the day here is a service that needs to query a knowledge base. Previously, probably at the time the exam was conceived, there was still the QnA service and also the Language service. You were probably in a transition. In any case The bot is the service that the user can use to query a knowledge base and get a response to their query. Today QnA no longer exists, and also Language service has undergone further change and omni-comprises LUIS, QnA and Text Analytics. So I think the answer given is correct. https://learn.microsoft.com/en-us/azure/bot-service/bot-service-overview?view=azure-bot-service-4.0 https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/overview P.S.: I re-did the same question to ChatGPT and now he also answers Azure Bot Service :-)

8. Question 8 of 25

   DRAG DROP - Match the types of AI workloads to the appropriate scenarios. To answer, drag the appropriate workload type from the column on the left to its scenario on the right. Each workload type may be used once, more than once, or not at all. NOTE: Each correct selection is worth one point.

   

   This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

   Show the answer figure

   Answer

   Explanation

   1. Computer vision 2. Natural language processing 3. Anomaly detection 4. Machine learning (Clustering) https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-ocr OCR or Optical Character Recognition is also referred to as text recognition or text extraction. Machine-learning based OCR techniques allow you to extract printed or handwritten text from images, such as posters, street signs and product labels, as well as from documents like articles, reports, forms, and invoices. The text is typically extracted as words, text lines, and paragraphs or text blocks, enabling access to digital version of the scanned text. This eliminates or significantly reduces the need for manual data entry.

   Adapted from community discussion by zellck · 2024-06-23 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn

   Answer figure

   

   View community discussion3

   Candidate context

   Question 8 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. Z

      zellck

      2024-06-23

      4 upvotes

      1. Computer vision 2. Natural language processing 3. Anomaly detection 4. Machine learning (Clustering) https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-ocr OCR or Optical Character Recognition is also referred to as text recognition or text extraction. Machine-learning based OCR techniques allow you to extract printed or handwritten text from images, such as posters, street signs and product labels, as well as from documents like articles, reports, forms, and invoices. The text is typically extracted as words, text lines, and paragraphs or text blocks, enabling access to digital version of the scanned text. This eliminates or significantly reduces the need for manual data entry.

   2. Z

      zellck

      2024-06-23

      1 upvote

      https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/sentiment-opinion-mining/overview#sentiment-analysis The sentiment analysis feature provides sentiment labels (such as "negative", "neutral" and "positive") based on the highest confidence score found by the service at a sentence and document-level. This feature also returns confidence scores between 0 and 1 for each document & sentences within it for positive, neutral and negative sentiment. https://learn.microsoft.com/en-us/training/modules/create-clustering-model-azure-machine-learning-designer/2-clustering-scenarios Clustering is a form of machine learning that is used to group similar items into clusters based on their features. For example, a researcher might take measurements of penguins, and group them based on similarities in their proportions.

   3. Z

      zellck

      2024-06-24

      1 upvote

      https://learn.microsoft.com/en-us/azure/cognitive-services/anomaly-detector/overview Anomaly Detector is an AI service with a set of APIs, which enables you to monitor and detect anomalies in your time series data with little machine learning (ML) knowledge, either batch validation or real-time inference.

9. Question 9 of 25

   You run a charity event that involves posting photos of people wearing sunglasses on Twitter. You need to ensure that you only retweet photos that meet the following requirements: ✑ Include one or more faces. ✑ Contain at least one person wearing sunglasses. What should you use to analyze the images?

   • Athe Verify operation in the Face service
   • Bthe Detect operation in the Face service
   • Cthe Describe Image operation in the Computer Vision service
   • Dthe Analyze Image operation in the Computer Vision service

   Show answer

   Correct answer: B

   How others answered

   • B100%

   Explanation

   B is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-identity#face-detection-and-analysis Face detection is required as a first step in all the other scenarios. The Detect API detects human faces in an image and returns the rectangle coordinates of their locations. It also returns a unique ID that represents the stored face data. This is used in later operations to identify or verify faces. Optionally, face detection can extract a set of face-related attributes, such as head pose, age, emotion, facial hair, and glasses. These attributes are general predictions, not actual classifications. Some attributes are useful to ensure that your application is getting high-quality face data when users add themselves to a Face service. For example, your application could advise users to take off their sunglasses if they're wearing sunglasses.

   Adapted from community discussion by zellck · 2023-06-25 · 5 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn
   • Microsoft Learn

   View community discussion4

   Candidate context

   Question 9 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

   1. I

      ipindado2020

      2020-12-03

      28 upvotes

      answer si correct https://docs.microsoft.com/en-us/azure/cognitive-services/face/concepts/face-detection Face detect can be requested to detect also glasses attribute

   2. P

      profesorklaus

      2022-03-19

      9 upvotes

      "Optionally, face detection can extract a set of face-related attributes, such as head pose, age, emotion, facial hair, and glasses." https://docs.microsoft.com/en-us/azure/cognitive-services/face/overview

   3. Z

      zellck

      2023-06-25

      Selected B5 upvotes

      B is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-identity#face-detection-and-analysis Face detection is required as a first step in all the other scenarios. The Detect API detects human faces in an image and returns the rectangle coordinates of their locations. It also returns a unique ID that represents the stored face data. This is used in later operations to identify or verify faces. Optionally, face detection can extract a set of face-related attributes, such as head pose, age, emotion, facial hair, and glasses. These attributes are general predictions, not actual classifications. Some attributes are useful to ensure that your application is getting high-quality face data when users add themselves to a Face service. For example, your application could advise users to take off their sunglasses if they're wearing sunglasses.

   4. V

      vchirag

      2024-05-15

      4 upvotes

      Computer Vision covers broader image analysis, the Face Service focuses specifically on human faces and offers more specialized features for face-related tasks.

10. Question 10 of 25

    When you design an AI system to assess whether loans should be approved, the factors used to make the decision should be explainable. This is an example of which Microsoft guiding principle for responsible AI?

    • Atransparency
    • Binclusiveness
    • Cfairness
    • Dprivacy and security

    Show answer

    Correct answer: A

    How others answered

    • A95%

    Explanation

    A is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#transparency Achieving transparency helps the team to understand the data and algorithms used to train the model, what transformation logic was applied to the data, the final model generated, and its associated assets. This information offers insights about how the model was created, which allows it to be reproduced in a transparent way. Snapshots within Azure Machine Learning workspaces support transparency by recording or retraining all training-related assets and metrics involved in the experiment.

    Adapted from community discussion by zellck · 2024-09-19 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 10 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. R

       Rezaphp

       2021-12-26

       4 upvotes

       Choice c (Fairness) is also correct to some extent. To make sure the models are unbiased toward some applicants characteristics such as gender, race, ..., we should be able to interpret the importance of feature on the model predictions as highlighted in the learning material for the exam as one of the solutions to ensure fairness, "Azure Machine Learning includes the capability to interpret models and quantify the extent to which each feature of the data influences the model's prediction." https://docs.microsoft.com/en-us/learn/modules/get-started-ai-fundamentals/8-understand-responsible-ai

    2. D

       DWTF

       2022-01-01

       5 upvotes

       The keyword in the question is "explainable", therefore, in that context, the answer should be "transparency".

    3. Z

       zellck

       2024-09-19

       Selected A2 upvotes

       A is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#transparency Achieving transparency helps the team to understand the data and algorithms used to train the model, what transformation logic was applied to the data, the final model generated, and its associated assets. This information offers insights about how the model was created, which allows it to be reproduced in a transparent way. Snapshots within Azure Machine Learning workspaces support transparency by recording or retraining all training-related assets and metrics involved in the experiment.

    4. R

       rdemontis

       2023-06-12

       Selected A1 upvote

       Provided answer is correct. Transparency is one of the guiding principles for responsible AI. It emphasizes the importance of making AI systems explainable and understandable to users and stakeholders. In the given scenario, designing an AI system to assess loan approvals and ensuring that the factors used in the decision-making process are explainable aligns with the principle of transparency. By providing transparency, users and stakeholders can understand how the AI system makes decisions and trust its outcomes.(Chat GPT)

11. Question 11 of 25

    You need to create a model that labels a collection of your personal digital photographs. Which Azure Cognitive Services service should you use?

    • AForm Recognizer
    • BCustom Vision
    • CLanguage
    • DComputer Vision

    Show answer

    Correct answer: B

    How others answered

    • B97%

    Explanation

    According to Microsoft's documentation, you can use Custom Vision to label a collection of your personal digital photographs ¹. With Custom Vision, you can upload your own images and labels, and train a model to recognize specific objects or patterns [^web page^]. You can also use Smart Labeler to generate suggested tags for images, which lets you label a large number of images more quickly when you're training a Custom Vision model ¹. Source: Conversation with Bing, 4/26/2023 (1) Label images faster with Smart Labeler - Azure Cognitive Services .... https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/suggested-tags Accessed 4/26/2023.

    Adapted from community discussion by master_yoda · 2023-04-26 · 12 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 11 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. M

       master_yoda

       2023-04-26

       Selected B12 upvotes

       According to Microsoft's documentation, you can use Custom Vision to label a collection of your personal digital photographs ¹. With Custom Vision, you can upload your own images and labels, and train a model to recognize specific objects or patterns [^web page^]. You can also use Smart Labeler to generate suggested tags for images, which lets you label a large number of images more quickly when you're training a Custom Vision model ¹. Source: Conversation with Bing, 4/26/2023 (1) Label images faster with Smart Labeler - Azure Cognitive Services .... https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/suggested-tags Accessed 4/26/2023.

    2. C

       CADMAN

       2024-01-08

       2 upvotes

       The appropriate Azure Cognitive Service to use for labeling a collection of personal digital photographs is D. Computer Vision. Azure Computer Vision provides pre-built models that can be used to analyze images and extract information about their content, including the ability to classify images into predefined categories. It offers a "Tags" feature that can automatically generate a list of descriptive words or phrases that represent the content of an image. Form Recognizer (option A) is designed for extracting structured data from forms and documents, while Custom Vision (option B) is for training custom models to classify images into user-defined categories. Language (option C) is for analyzing and processing text data, not images. Therefore, Azure Computer Vision is the most suitable service for labeling a collection of personal digital photographs by automatically generating descriptive tags for the images.

    3. R

       rdemontis

       2023-06-09

       Selected B4 upvotes

       given answer is wrong. Only with custom vision you can train a model using your images and tag. Computer Vision is not trainable or modifiable. It's a ready to go package.

    4. Z

       zellck

       2023-06-23

       Selected B2 upvotes

       B is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/overview Azure Custom Vision is an image recognition service that lets you build, deploy, and improve your own image identifier models. An image identifier applies labels to images, according to their visual characteristics. Each label represents a classification or object. Unlike the Computer Vision service, Custom Vision allows you to specify your own labels and train custom models to detect them.

12. Question 12 of 25

    Your company manufactures widgets. You have 1,000 digital photos of the widgets. You need to identify the location of the widgets within the photos. What should you use?

    • AComputer Vision Spatial Analysis
    • BCustom Vision object detection
    • CComputer Vision Image Analysis
    • DCustom Vision classification

    Show answer

    Correct answer: B

    How others answered

    • B100%

    Explanation

    correct. Custom Vision object detection is a service that allows you to train a model to detect and locate specific objects within images. In this case, you can train a Custom Vision object detection model using the 1,000 digital photos of the widgets. By labeling the widgets in the images and providing corresponding bounding boxes, the model can learn to identify and locate the widgets in new photos. Computer Vision Spatial Analysis focuses on analyzing the spatial relationships and layout of objects within an image, which might not be suitable for precisely identifying the location of specific objects like widgets. Computer Vision Image Analysis is a broad term that encompasses various image analysis techniques, but it may not provide the specific object detection capability required for locating widgets. Custom Vision classification is used to classify images into predefined categories rather than detecting the location of specific objects within the images.

    Adapted from community discussion by rdemontis · 2024-06-09 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 12 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. R

       rdemontis

       2024-06-09

       Selected B4 upvotes

       correct. Custom Vision object detection is a service that allows you to train a model to detect and locate specific objects within images. In this case, you can train a Custom Vision object detection model using the 1,000 digital photos of the widgets. By labeling the widgets in the images and providing corresponding bounding boxes, the model can learn to identify and locate the widgets in new photos. Computer Vision Spatial Analysis focuses on analyzing the spatial relationships and layout of objects within an image, which might not be suitable for precisely identifying the location of specific objects like widgets. Computer Vision Image Analysis is a broad term that encompasses various image analysis techniques, but it may not provide the specific object detection capability required for locating widgets. Custom Vision classification is used to classify images into predefined categories rather than detecting the location of specific objects within the images.

    2. Z

       zellck

       2024-06-23

       Selected B2 upvotes

       B is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/overview Azure Custom Vision is an image recognition service that lets you build, deploy, and improve your own image identifier models. An image identifier applies labels to images, according to their visual characteristics. Each label represents a classification or object. Unlike the Computer Vision service, Custom Vision allows you to specify your own labels and train custom models to detect them.

    3. Z

       zellck

       2024-06-23

       2 upvotes

       https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/concept-object-detection Object detection is similar to tagging, but the API returns the bounding box coordinates (in pixels) for each object found in the image. For example, if an image contains a dog, cat and person, the Detect operation will list those objects with their coordinates in the image. You can use this functionality to process the relationships between the objects in an image. It also lets you determine whether there are multiple instances of the same object in an image.

    4. P

       PeteColag

       2024-11-19

       2 upvotes

       You need to use Custom Vision in order to train the model to recognize the widge.

13. Question 13 of 25

    HOTSPOT - To complete the sentence, select the appropriate option in the answer area.

    

    This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

    Show the answer figure

    Answer

    Explanation

    It doesnt really matter which it is as Microsoft discontinues these parts of concept face detection. Question therefore outdated. "Microsoft will be retiring facial recognition capabilities that can be used to try to infer emotional states and identity attributes which, if misused, can subject people to stereotyping, discrimination or unfair denial of services. These include capabilities that predict emotion, gender, age, smile, facial hair, hair and makeup. Existing customers have until June 30, 2023 to discontinue use of these capabilities before they are retired. " https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/concept-face-detection

    Adapted from community discussion by AzlearnRB · 2023-07-16 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    Answer figure

    

    View community discussion4

    Candidate context

    Question 13 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. A

       AzlearnRB

       2023-07-16

       4 upvotes

       It doesnt really matter which it is as Microsoft discontinues these parts of concept face detection. Question therefore outdated. "Microsoft will be retiring facial recognition capabilities that can be used to try to infer emotional states and identity attributes which, if misused, can subject people to stereotyping, discrimination or unfair denial of services. These include capabilities that predict emotion, gender, age, smile, facial hair, hair and makeup. Existing customers have until June 30, 2023 to discontinue use of these capabilities before they are retired. " https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/concept-face-detection

    2. R

       rumenta

       2023-12-19

       6 upvotes

       Detection https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-face-detection#attributes

    3. Z

       zellck

       2023-06-23

       5 upvotes

       Same as Question 186. https://www.examtopics.com/discussions/microsoft/view/100263-exam-ai-900-topic-1-question-186-discussion

    4. R

       rdemontis

       2023-06-09

       4 upvotes

       I think facial analysis is correct in this question because the Face service, although it has now evolved and includes it in Detection, used to have a specific facial analysis service as part of the Face API. And I don't think this question is that up-to-date, otherwise it wouldn't even be there among the Facial Analysis answers.

14. Question 14 of 25

    DRAG DROP - Match the Azure Cognitive Services to the appropriate AI workloads. To answer, drag the appropriate service from the column on the left to its workload on the right. Each service may be used once, more than once, or not at all. NOTE: Each correct match is worth one point.

    

    This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

    Show the answer figure

    Answer

    Explanation

    https://learn.microsoft.com/en-us/azure/applied-ai-services/form-recognizer/overview?view=form-recog-3.0.0 Azure Form Recognizer is a cloud-based Azure Applied AI Service that enables you to build intelligent document processing solutions. Massive amounts of data, spanning a wide variety of data types, are stored in forms and documents. Form Recognizer enables you to effectively manage the velocity at which data is collected and processed and is key to improved operations, informed data-driven decisions, and enlightened innovation. https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-identity The Azure Face service provides AI algorithms that detect, recognize, and analyze human faces in images. Facial recognition software is important in many different scenarios, such as identity verification, touchless access control, and face blurring for privacy.

    Adapted from community discussion by zellck · 2023-06-23 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    Answer figure

    

    View community discussion4

    Candidate context

    Question 14 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. Z

       zellck

       2023-06-23

       2 upvotes

       https://learn.microsoft.com/en-us/azure/applied-ai-services/form-recognizer/overview?view=form-recog-3.0.0 Azure Form Recognizer is a cloud-based Azure Applied AI Service that enables you to build intelligent document processing solutions. Massive amounts of data, spanning a wide variety of data types, are stored in forms and documents. Form Recognizer enables you to effectively manage the velocity at which data is collected and processed and is key to improved operations, informed data-driven decisions, and enlightened innovation. https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/overview-identity The Azure Face service provides AI algorithms that detect, recognize, and analyze human faces in images. Facial recognition software is important in many different scenarios, such as identity verification, touchless access control, and face blurring for privacy.

    2. R

       Rosviul

       2023-03-21

       4 upvotes

       repeated- exactly the same as Q185: answer: custom vision - form recognizer - face

    3. Z

       zellck

       2023-06-23

       3 upvotes

       Same as Question 185. https://www.examtopics.com/discussions/microsoft/view/106116-exam-ai-900-topic-1-question-185-discussion

    4. Z

       zellck

       2023-06-23

       1 upvote

       1. Custom Vision 2. Form Recognized 3. Face https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/overview Azure Custom Vision is an image recognition service that lets you build, deploy, and improve your own image identifier models. An image identifier applies labels to images, according to their visual characteristics. Each label represents a classification or object. Unlike the Computer Vision service, Custom Vision allows you to specify your own labels and train custom models to detect them.

15. Question 15 of 25

    Natural language processing can be used to __________. Select the answer that correctly completes the sentence.

    • AAnalyze video content
    • BGenerate speech
    • CClassify email messages as work-related or personal.
    • DClassify images

    Show answer

    Correct answer: C

    How others answered

    • C100%

    Explanation

    I think B is also a correct answer, check the Microsoft AI fundamental self-learning web site Recognize and synthesize speech with Azure AI Speech (https://learn.microsoft.com/en-us/training/modules/recognize-synthesize-speech/) this section is under Explore natural language processing training module (https://learn.microsoft.com/en-us/training/paths/explore-natural-language-processing/)

    Adapted from community discussion by vlu · 2023-09-14 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion3

    Candidate context

    Question 15 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. B

       BobFar

       2024-07-24

       1 upvote

       C. Classify email messages as work-related or personal. Explanation: Natural Language Processing (NLP) is a field of artificial intelligence that focuses on the interaction between computers and humans through natural language. Here are the reasons why each option is appropriate or not: A. Analyze video content: This is not correct because analyzing video content typically involves computer vision techniques, not NLP. B. Generate speech: While generating speech can be related to NLP (e.g., text-to-speech systems), it is more directly associated with speech synthesis and not purely NLP. C. Classify email messages as work-related or personal: This is correct. NLP can be used to analyze the text content of emails and classify them based on their context, making it an appropriate example of an NLP application. D. Classify images: This is not correct because classifying images typically involves image processing and computer vision techniques, not NLP. https://learn.microsoft.com/en-us/azure/ai-services/language-service/custom-text-classification/overview

    2. V

       vlu

       2023-09-14

       3 upvotes

       I think B is also a correct answer, check the Microsoft AI fundamental self-learning web site Recognize and synthesize speech with Azure AI Speech (https://learn.microsoft.com/en-us/training/modules/recognize-synthesize-speech/) this section is under Explore natural language processing training module (https://learn.microsoft.com/en-us/training/paths/explore-natural-language-processing/)

    3. R

       rdemontis

       2023-06-15

       Selected C1 upvote

       Natural language processing (NLP) involves the use of algorithms and techniques to understand and analyze human language. It enables computers to process and interpret text or speech data. In the given options, classifying email messages as work-related or personal is a task that falls under the domain of NLP. NLP techniques can be used to extract relevant information from text data and categorize it based on different criteria, such as topic, sentiment, intent, or in this case, distinguishing between work-related and personal emails. (chatGPT)

16. Question 16 of 25

    An app that analyzes social media posts to identify their tone is an example of which type of natural language processing (NLP) workload?

    • Asentiment analysis
    • Bspeech recognition
    • Ckey phrase extraction
    • Dentity recognition

    Show answer

    Correct answer: A

    How others answered

    • A100%

    Explanation

    An app that analyzes social media posts to identify their tone is an example of a sentiment analysis natural language processing (NLP) workload.

    Adapted from community discussion by XtraWest · 2023-04-08 · 6 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 16 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. X

       XtraWest

       2023-04-08

       6 upvotes

       An app that analyzes social media posts to identify their tone is an example of a sentiment analysis natural language processing (NLP) workload.

    2. Z

       zellck

       2023-06-11

       Selected A1 upvote

       A is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/sentiment-opinion-mining/overview Sentiment analysis and opinion mining are features offered by Azure Cognitive Service for Language, a collection of machine learning and AI algorithms in the cloud for developing intelligent applications that involve written language. These features help you find out what people think of your brand or topic by mining text for clues about positive or negative sentiment, and can associate them with specific aspects of the text.

    3. Z

       zellck

       2023-06-11

       1 upvote

       https://learn.microsoft.com/en-us/training/modules/analyze-text-with-text-analytics-service/2-get-started-azure - Sentiment analysis The text analytics capabilities in the Language service can evaluate text and return sentiment scores and labels for each sentence. This capability is useful for detecting positive and negative sentiment in social media, customer reviews, discussion forums and more.

    4. B

       BobFar

       2024-07-24

       1 upvote

       A. sentiment analysis is correct answer! https://learn.microsoft.com/en-us/azure/ai-services/language-service/sentiment-opinion-mining/overview?tabs=prebuilt

17. Question 17 of 25

    You have a solution that analyzes social media posts to extract the mentions of city names and the city names discussed most frequently. Which type of natural language processing (NLP) workload does the solution use?

    • Aspeech recognition
    • Bsentiment analysis
    • Ckey phrase extraction
    • Dentity recognition

    Show answer

    Correct answer: D

    How others answered

    • D100%

    Explanation

    D is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/named-entity-recognition/overview Named Entity Recognition (NER) is one of the features offered by Azure Cognitive Service for Language, a collection of machine learning and AI algorithms in the cloud for developing intelligent applications that involve written language. The NER feature can identify and categorize entities in unstructured text. For example: people, places, organizations, and quantities.

    Adapted from community discussion by zellck · 2024-12-11 · 6 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 17 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. Z

       zellck

       2024-12-11

       Selected D6 upvotes

       D is the answer. https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/named-entity-recognition/overview Named Entity Recognition (NER) is one of the features offered by Azure Cognitive Service for Language, a collection of machine learning and AI algorithms in the cloud for developing intelligent applications that involve written language. The NER feature can identify and categorize entities in unstructured text. For example: people, places, organizations, and quantities.

    2. B

       Brandon_Marlin

       2024-12-04

       Selected D3 upvotes

       The correct answer is D. entity recognition. Entity recognition is a natural language processing (NLP) task that involves identifying and extracting named entities from text. Named entities are typically people, places, organizations, dates, and times. In this case, the solution is extracting the mentions of city names, which are places. The other options are not correct. Speech recognition is the process of converting spoken language into text. Sentiment analysis is the process of identifying the sentiment of a text, such as whether it is positive, negative, or neutral. Key phrase extraction is the process of identifying the most important words or phrases in a text.

    3. Z

       zellck

       2024-12-11

       3 upvotes

       https://learn.microsoft.com/en-us/training/modules/analyze-text-with-text-analytics-service/2-get-started-azure - Entity recognition You can provide the Language service with unstructured text and it will return a list of entities in the text that it recognizes. The service can also provide links to more information about that entity on the web. An entity is essentially an item of a particular type or a category; and in some cases, subtype, such as those as shown in the following table.

    4. F

       fluffybytes

       2024-11-19

       Selected D3 upvotes

       Mentioning city names and locations in a text is entity recognition, not key phrase extraction.

18. Question 18 of 25

    Select 2

    You are building an AI-based app. You need to ensure that the app uses the principles for responsible AI. Which two principles should you follow? Each correct answer presents part of the solution. NOTE: Each correct selection is worth one point.

    • AImplement an Agile software development methodology
    • BImplement a process of AI model validation as part of the software review process
    • CEstablish a risk governance committee that includes members of the legal team, members of the risk management team, and a privacy officer
    • DPrevent the disclosure of the use of AI-based algorithms for automated decision making

    Show answer

    Correct answer: B, C

    How others answered

    • BC100%

    Explanation

    BC is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#reliability-and-safety AI systems need to be reliable and safe in order to be trusted. It's important for a system to perform as it was originally designed and for it to respond safely to new situations. Its inherent resilience should resist intended or unintended manipulation. Rigorous testing and validation should be established for operating conditions to ensure that the system responds safely to edge cases, and A/B testing and champion/challenger methods should be integrated into the evaluation process.

    Adapted from community discussion by zellck · 2023-06-24 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 18 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. M

       Marskmen

       2021-11-02

       22 upvotes

       B and C are correct. B ensures reliability and safety principle and C ensures privacy and security principle of AI.

    2. Z

       zellck

       2023-06-24

       Selected B, C3 upvotes

       BC is the answer. https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#reliability-and-safety AI systems need to be reliable and safe in order to be trusted. It's important for a system to perform as it was originally designed and for it to respond safely to new situations. Its inherent resilience should resist intended or unintended manipulation. Rigorous testing and validation should be established for operating conditions to ensure that the system responds safely to edge cases, and A/B testing and champion/challenger methods should be integrated into the evaluation process.

    3. Z

       zellck

       2023-06-24

       3 upvotes

       https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/innovate/best-practices/trusted-ai#accountability Accountability is an essential pillar of responsible AI. The people who design and deploy the AI system need to be accountable for its actions and decisions, especially as we progress toward more autonomous systems. Organizations should consider establishing an internal review body that provides oversight, insights, and guidance about developing and deploying AI systems. While this guidance might vary depending on the company and region, it should reflect an organization's AI journey.

    4. R

       rdemontis

       2023-05-29

       Selected B, C3 upvotes

       correct answer. B --> Safety and reliability, C --> Privacy and security

19. Question 19 of 25

    You have the Predicted vs. True chart shown in the following exhibit.

    

    Which type of model is the chart used to evaluate?

    • Aclassification
    • Bregression
    • Cclustering

    Show answer

    Correct answer: B

    How others answered

    • B100%

    Explanation

    B is the answer. https://learn.microsoft.com/en-us/training/modules/create-regression-model-azure-machine-learning-designer/2-regression-scenarios Regression is a form of machine learning used to understand the relationships between variables to predict a desired outcome. Regression predicts a numeric label or outcome based on variables, or features. For example, an automobile sales company might use the characteristics of a car (such as engine size, number of seats, mileage, and so on) to predict its likely selling price. In this case, the characteristics of the car are the features, and the selling price is the label.

    Adapted from community discussion by zellck · 2023-06-25 · 9 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 19 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. Z

       zellck

       2023-06-25

       Selected B9 upvotes

       B is the answer. https://learn.microsoft.com/en-us/training/modules/create-regression-model-azure-machine-learning-designer/2-regression-scenarios Regression is a form of machine learning used to understand the relationships between variables to predict a desired outcome. Regression predicts a numeric label or outcome based on variables, or features. For example, an automobile sales company might use the characteristics of a car (such as engine size, number of seats, mileage, and so on) to predict its likely selling price. In this case, the characteristics of the car are the features, and the selling price is the label.

    2. E

       endeesa

       2023-03-29

       Selected B4 upvotes

       We are predicting a continuous numerical variable so answer must be Regression

    3. S

       Sylph

       2023-08-16

       4 upvotes

       Correct https://learn.microsoft.com/en-us/azure/machine-learning/how-to-understand-automated-ml?view=azureml-api-2#predicted-vs-true

    4. S

       SnottyPudding

       2021-09-19

       3 upvotes

       Corrected link: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-understand-automated-ml

20. Question 20 of 25

    Which metric can you use to evaluate a classification model?

    • Atrue positive rate
    • Bmean absolute error (MAE)
    • Ccoefficient of determination (R2)
    • Droot mean squared error (RMSE)

    Show answer

    Correct answer: A

    How others answered

    • A100%

    Explanation

    FYI: the concept Receiver Operating Characteristic (ROC) was developed during World War II for the analysis of radar receivers. Reference: https://stats.stackexchange.com/questions/341043/what-is-the-origin-of-the-receiver-operating-characteristic-roc-terminology

    Adapted from community discussion by ssshhhh · 2023-03-22 · 12 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 20 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. G

       GetulioJr

       2022-12-03

       27 upvotes

       MAE, RMSE and R2 are metris for regression: https://docs.microsoft.com/en-us/azure/machine-learning/algorithm-module-reference/evaluate-model So answer is correct.

    2. S

       ssshhhh

       2023-03-22

       12 upvotes

       FYI: the concept Receiver Operating Characteristic (ROC) was developed during World War II for the analysis of radar receivers. Reference: https://stats.stackexchange.com/questions/341043/what-is-the-origin-of-the-receiver-operating-characteristic-roc-terminology

    3. T

       Tanat

       2023-01-04

       3 upvotes

       "A" is the correct answer https://docs.microsoft.com/en-us/azure/machine-learning/how-to-understand-automated-ml#classification

    4. E

       ExamTopics007

       2022-06-01

       1 upvote

       AUC is the Area under the Receiver Operating Characteristic Curve. Micro is computed globally by combining the true positives and false positives from each class.

21. Question 21 of 25

    Select 2

    Which two components can you drag onto a canvas in Azure Machine Learning designer? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

    • Adataset
    • Bcompute
    • Cpipeline
    • Dmodule

    Show answer

    Correct answer: A, D

    How others answered

    • AD100%

    Explanation

    Azure Machine Learning designer lets you visually connect datasets and modules on an interactive canvas to create machine learning models.

    Adapted from community discussion by ExamTopics007 · 2021-06-01 · 49 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn
    • azure.microsoft.com documentation
    • azure.microsoft.com documentation

    View community discussion4

    Candidate context

    Question 21 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. E

       ExamTopics007

       2021-06-01

       49 upvotes

       Azure Machine Learning designer lets you visually connect datasets and modules on an interactive canvas to create machine learning models.

    2. M

       Mehe323

       2024-07-06

       9 upvotes

       The audacity to call other people sheepS (tip: sheep is both singular and plural) and having the answer wrong... https://azure.microsoft.com/en-au/products/machine-learning/designer/ https://azure.microsoft.com/en-au/products/machine-learning/designer

    3. N

       nohaph

       2021-06-28

       4 upvotes

       The designer gives you a visual canvas to build, test, and deploy machine learning models. With the designer you can: Drag-and-drop datasets and modules onto the canvas.

    4. M

       mfreak

       2024-03-24

       3 upvotes

       A "module" refers to an individual operation or step that you can add to your machine learning workflow. These modules represent specific tasks or operations, such as data preprocessing, feature engineering, model training, model scoring, and evaluation.

22. Question 22 of 25

    Select 2

    Which two languages can you use to write custom code for Azure Machine Learning designer? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

    • APython
    • BR
    • CC#
    • DScala

    Show answer

    Correct answer: A, B

    How others answered

    • AB100%

    Explanation

    "Build and train machine learning models with state-of-the art machine learning and deep learning algorithms, including those for computer vision, text analytics, recommendations, and anomaly detection. Drag and drop modules for no-code models or customize using Python and R code." as per the link given: https://azure.microsoft.com/en-us/products/machine-learning/designer/#features

    Adapted from community discussion by arska · 2023-04-05 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • azure.microsoft.com documentation
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 22 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. A

       arska

       2023-04-05

       Selected A, B3 upvotes

       "Build and train machine learning models with state-of-the art machine learning and deep learning algorithms, including those for computer vision, text analytics, recommendations, and anomaly detection. Drag and drop modules for no-code models or customize using Python and R code." as per the link given: https://azure.microsoft.com/en-us/products/machine-learning/designer/#features

    2. Z

       zellck

       2023-12-25

       Selected A, B3 upvotes

       AB is the answer. https://azure.microsoft.com/en-us/products/machine-learning/designer/#features Build and train machine learning models with state-of-the art machine learning and deep learning algorithms, including those for computer vision, text analytics, recommendations, and anomaly detection. Drag and drop modules for no-code models or customize using Python and R code.

    3. A

       Alejandre

       2023-04-20

       3 upvotes

       Phyton and R: https://learn.microsoft.com/en-us/azure/architecture/data-guide/technology-choices/data-science-and-machine-learning

    4. R

       rdemontis

       2023-11-30

       Selected A, B2 upvotes

       correct answer. https://www.pluralsight.com/guides/r-and-python-scripts-in-azure-ml-studio

23. Question 23 of 25

    Your company wants to build a recycling machine for bottles. The recycling machine must automatically identify bottles of the correct shape and reject all other items. Which type of AI workload should the company use?

    • Aanomaly detection
    • Bconversational AI
    • Ccomputer vision
    • Dnatural language processing

    Show answer

    Correct answer: C

    How others answered

    • C100%

    Explanation

    C is correct. Anomaly detection is unsupervised learning that detect "deviation" from norm. To correctly identify the "correct" plastic bottle, computer vision should be used.

    Adapted from community discussion by fatrapb1t · 2021-05-23 · 51 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 23 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. F

       fatrapb1t

       2021-05-23

       51 upvotes

       C is correct. Anomaly detection is unsupervised learning that detect "deviation" from norm. To correctly identify the "correct" plastic bottle, computer vision should be used.

    2. R

       rdemontis

       2023-11-30

       8 upvotes

       Computer vision focuses on teaching machines to interpret and understand visual information from images or videos. In this case, the recycling machine needs to analyze the shape of items to determine whether they are bottles or not. By using computer vision techniques, such as image classification or object detection, the machine can process visual inputs and accurately identify bottles based on their shape. Anomaly detection, conversational AI, and natural language processing are not directly applicable to this scenario: Anomaly detection is used to identify rare or unusual patterns or outliers in data, which is not the primary requirement for the recycling machine. Conversational AI involves building conversational agents or chatbots to interact with users using natural language, which is not the main objective of the recycling machine. Natural language processing focuses on processing and understanding human language, which is not directly relevant for identifying bottles based on their shape.

    3. J

       Johnnien

       2021-06-12

       10 upvotes

       Anomaly detection might indicate fraud but is not used to identify the correct bottle shape.

    4. E

       ET2020

       2021-07-05

       9 upvotes

       The right answer is C: Computer Vision. You can train an Image Classification model to "identify bottles of the correct shape and reject all other items".

24. Question 24 of 25

    DRAG DROP - Match the facial recognition tasks to the appropriate questions. To answer, drag the appropriate task from the column on the left to its question on the right. Each task may be used once, more than once, or not at all. NOTE: Each correct selection is worth one point. Select and Place:

    

    This is an interactive task on the real exam — a drag-and-drop, hotspot or ordering question. There is nothing to select here, so the answer is shown below for reference.

    Show the answer figure

    Answer

    Explanation

    Box 1: verification - Face verification: Check the likelihood that two faces belong to the same person and receive a confidence score. Box 2: similarity - Box 3: Grouping - Box 4: identification - Face detection: Detect one or more human faces along with attributes such as: age, emotion, pose, smile, and facial hair, including 27 landmarks for each face in the image. Reference: https://azure.microsoft.com/en-us/services/cognitive-services/face/#features

    Adapted from community discussion by idioteque · 2022-08-14 · 30 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    Answer figure

    

    View community discussion4

    Candidate context

    Question 24 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. I

       idioteque

       2022-08-14

       30 upvotes

       For me to remember :) verification = same person? similarity = look like? grouping = belong together? identification = who is this person?

    2. G

       gargaditya

       2020-10-30

       10 upvotes

       ~Face verification: The Verify API does an authentication against two detected faces or from one detected face to one person object. Practically, it evaluates whether two faces belong to the same person. ~Person identification: The Identify API is used to identify a detected face against a database of people (facial recognition search). This feature might be useful for automatic image tagging in photo management software. You create the database in advance, and you can edit it over time.

    3. G

       gargaditya

       2020-10-30

       4 upvotes

       ~Find similar faces: The Find Similar API does face matching between target face and a set of candidate faces, finding a smaller set of faces that look similar to the target face. This is useful for doing a face search by image. Two working modes, matchPerson and matchFace, are supported. The matchPerson mode returns similar faces after filtering for the same person by using the Verify API. The matchFace mode ignores the same-person filter. It returns a list of similar candidate faces that may or may not belong to the same person. ~Face grouping: The Group API divides a set of unknown faces into several groups based on similarity. Each group is a disjoint proper subset of the original set of faces. All of the faces in a group are likely to belong to the same person. There can be several different groups for a single person. The groups are differentiated by another factor, such as expression, for example.

    4. S

       SnakePlissken

       2021-04-14

       4 upvotes

       Correct. Do two images of a face belong to the same person? This defines verification. Does this person look like other people? This defines similarity. Do all of these faces belong together? This defines grouping. Who is this person in this group of people? This defines identification. https://docs.microsoft.com/en-us/learn/modules/identify-faces-with-computer-vision/4-overview-of-face-recognition

25. Question 25 of 25

    Select 2

    What are two tasks that can be performed by using computer vision? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

    • APredict stock prices.
    • BDetect brands in an image.
    • CDetect the color scheme in an image
    • DTranslate text between languages.
    • EExtract key phrases.

    Show answer

    Correct answer: B, C

    How others answered

    • BC100%

    Explanation

    Detech Color scheme and Brand images are correct based on Detect the color scheme Analyze color usage within an image. Computer Vision can determine whether an image is black & white or color and, for color images, identify the dominant and accent colors. Detect the color scheme and Azure's Computer Vision service gives you access to advanced algorithms that process images and return information based on the visual features you're interested in. For example, Computer Vision can determine whether an image contains adult content, find specific brands or objects, or find human faces.

    Adapted from community discussion by wreck · 2022-04-04 · 25 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn
    • Microsoft Learn

    View community discussion4

    Candidate context

    Question 25 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by AI-900 or CertExamHQ.

    1. W

       wreck

       2022-04-04

       25 upvotes

       Detech Color scheme and Brand images are correct based on Detect the color scheme Analyze color usage within an image. Computer Vision can determine whether an image is black & white or color and, for color images, identify the dominant and accent colors. Detect the color scheme and Azure's Computer Vision service gives you access to advanced algorithms that process images and return information based on the visual features you're interested in. For example, Computer Vision can determine whether an image contains adult content, find specific brands or objects, or find human faces.

    2. I

       Ihueghian

       2022-11-03

       12 upvotes

       A. Predict stock prices. (ML Regression) B. Detect brands in an image. (Computer Vision) C. Detect the color scheme in an image (Computer Vision) D. Translate text between languages. (PNL, Translator Text) E. Extract key phrases. (AI Workload Computer Vision, as Azure service is Form Recognizer)

    3. [

       [Removed]

       2022-11-30

       8 upvotes

       Extracting key phrases, is not a computer vision, it is a part of NLP. B and C are the right answers

    4. G

       gargaditya

       2022-04-30

       4 upvotes

       Incorrect answers, correct ones are B and C https://docs.microsoft.com/en-in/learn/modules/analyze-images-computer-vision/2-image-analysis-azure Detecting brands This feature provides the ability to identify commercial brands. The service has an existing database of thousands of globally recognized logos from commercial brands of products Additional capabilities In addition to these capabilities, the Computer Vision service can: Detect image types - for example, identifying clip art images or line drawings. Detect image color schemes - specifically, identifying the dominant foreground, background, and overall colors in an image. Generate thumbnails - creating small versions of images. Moderate content - detecting images that contain adult content or depict violent, gory scenes.

Keyboard:J Kmove between questionsA–Dselect an answerRshow or hide the answer← →change page

AI-900 practice, question 1 of 25