Microsoft Microsoft Certifications AI-102 Questions & Answers

• Question 191:

  DRAG DROP

  You plan to use containerized versions of the Anomaly Detector API on local devices for testing and in on-premises datacenters.

  You need to ensure that the containerized deployments meet the following requirements:

  Prevent billing and API information from being stored in the command-line histories of the devices that run the container.

  Control access to the container images by using Azure role-based access control (Azure RBAC).

  Which four actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order. (Choose four.)

  NOTE: More than one order of answer choices is correct. You will receive credit for any of the correct orders you select.

  Select and Place:

  

  Correct Answer:

  

  Step 1: Pull the Anomaly Detector container image.

  Step 2: Create a custom Dockerfile

  Step 3: Build the image

  Step 4: Push the image to an Azure container registry.

  https://docs.microsoft.com/en-us/azure/cognitive-services/containers/container-reuse-recipe

• Question 192:

  DRAG DROP

  You are developing a call to the Face API. The call must find similar faces from an existing list named employeefaces. The employeefaces list contains 60,000 images.

  How should you complete the body of the HTTP request? To answer, drag the appropriate values to the correct targets. Each value 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.

  Select and Place:

  

  Correct Answer:

  

  Box 1: LargeFaceListID

  LargeFaceList: Add a face to a specified large face list, up to 1,000,000 faces.

  Note: Given query face's faceId, to search the similar-looking faces from a faceId array, a face list or a large face list. A "faceListId" is created by FaceList - Create containing persistedFaceIds that will not expire. And a "largeFaceListId" is

  created by LargeFaceList - Create containing persistedFaceIds that will also not expire.

  Incorrect Answers:

  Not "faceListId": Add a face to a specified face list, up to 1,000 faces.

  Box 2: matchFace

  Find similar has two working modes, "matchPerson" and "matchFace". "matchPerson" is the default mode that it tries to find faces of the same person as possible by using internal same-person thresholds. It is useful to find a known person's

  other photos. Note that an empty list will be returned if no faces pass the internal thresholds. "matchFace" mode ignores same-person thresholds and returns ranked similar faces anyway, even the similarity is low. It can be used in the cases

  like searching celebrity-looking faces.

  Reference:

  https://docs.microsoft.com/en-us/rest/api/faceapi/face/findsimilar

• Question 193:

  DRAG DROP

  You have 100 chatbots that each has its own Language Understanding model.

  Frequently, you must add the same phrases to each model.

  You need to programmatically update the Language Understanding models to include the new phrases.

  How should you complete the code? To answer, drag the appropriate values to the correct targets. Each value 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.

  Select and Place:

  

  Correct Answer:

  

  Box 1: AddPhraseListAsync Example: Add phraselist feature

  var phraselistId = await client.Features.AddPhraseListAsync(appId, versionId, new PhraselistCreateObject

  { EnabledForAllModels = false, IsExchangeable = true, Name = "QuantityPhraselist", Phrases = "few,more,extra"

  });

  Box 2: PhraselistCreateObject

  Reference: https://docs.microsoft.com/en-us/azure/cognitive-services/luis/client-libraries-rest-api

• Question 194:

  DRAG DROP

  You plan to use a Language Understanding application named app1 that is deployed to a container.

  App1 was developed by using a Language Understanding authoring resource named lu1.

  App1 has the versions shown in the following table.

  

  You need to create a container that uses the latest deployable version of app1.

  Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order. (Choose three.)

  Select and Place:

  

  Correct Answer:

  

• Question 195:

  You are building an app that will use the Speech service.

  You need to ensure that the app can authenticate to the service by using a Microsoft Azure Active Directory (Azure AD), part of Microsoft Entra, token.

  Which two actions should you perform? Each correct answer presents part of the solution.

  NOTE: Each correct selection is worth one point.

  A. Enable a virtual network service endpoint.

  B. Configure a custom subdomain.

  C. Request an X.509 certificate.

  D. Create a private endpoint.

  E. Create a Conditional Access policy.

  Correct Answer: BE

• Question 196:

  You have the following Python function for creating Azure Cognitive Services resources programmatically. def create_resource (resource_name, kind, account_tier, location) : parameters = CognitiveServicesAccount(sku=Sku

  (name=account_tier), kind=kind, location=location, properties={}) result = client.accounts.create(resource_group_name, resource_name, parameters)

  You need to call the function to create a free Azure resource in the West US Azure region. The resource will be used to generate captions of images automatically.

  Which code should you use?

  A. create_resource("res1", "ComputerVision", "F0", "westus")

  B. create_resource("res1", "CustomVision.Prediction", "F0", "westus")

  C. create_resource("res1", "ComputerVision", "S0", "westus")

  D. create_resource("res1", "CustomVision.Prediction", "S0", "westus")

  Correct Answer: A

• Question 197:

  You need to build a solution that will use optical character recognition (OCR) to scan sensitive documents by using the Computer Vision API. The solution must not be deployed to the public cloud. What should you do?

  A. Build an on-premises web app to query the Computer Vision endpoint.

  B. Host the Computer Vision endpoint in a container on an on-premises server.

  C. Host an exported Open Neural Network Exchange (ONNX) model on an on-premises server.

  D. Build an Azure web app to query the Computer Vision endpoint.

  Correct Answer: B

  One option to manage your Computer Vision containers on-premises is to use Kubernetes and Helm.

  Three primary parameters for all Cognitive Services containers are required. The Microsoft Software License Terms must be present with a value of accept. An Endpoint URI and API key are also needed.

  Incorrect:

  Not D: This Computer Vision endpoint would be available for the public, unless it is secured.

  Reference:

  https://docs.microsoft.com/en-us/azure/cognitive-services/computer-vision/deploy-computer-vision-on-premises

• Question 198:

  You have an Azure Cognitive Search solution and a collection of handwritten letters stored as JPEG files.

  You plan to index the collection. The solution must ensure that queries can be performed on the contents of the letters.

  You need to create an indexer that has a skillset.

  Which skill should you include?

  A. image analysis

  B. optical character recognition (OCR)

  C. key phrase extraction

  D. document extraction

  Correct Answer: B

  To ensure that queries can be performed on the contents of the letters, the skill that should be included in the indexer is optical character recognition (OCR).

  Option B, optical character recognition (OCR), is a technology that can recognize text within an image and convert it into machine-readable text. This skill will enable the search engine to read the handwritten letters and convert them into

  searchable text that can be indexed by Azure Cognitive Search.

  Option A, image analysis, is a useful skill for analyzing images to extract metadata, but it does not directly enable text recognition.

  Option C, key phrase extraction, extracts important phrases and concepts from text, but it requires the text to be already recognized and extracted by OCR or other text extraction techniques.

  Option D, document extraction, is a skill that extracts specific pieces of information from documents, but it does not address the challenge of recognizing and extracting text from handwritten letters.

• Question 199:

  You have an app that captures live video of exam candidates.

  You need to use the Face service to validate that the subjects of the videos are real people.

  What should you do?

  A. Call the face detection API and retrieve the face rectangle by using the FaceRectangle attribute.

  B. Call the face detection API repeatedly and check for changes to the FaceAttributes.HeadPose attribute.

  C. Call the face detection API and use the FaceLandmarks attribute to calculate the distance between pupils.

  D. Call the face detection API repeatedly and check for changes to the FaceAttributes.Accessories attribute.

  Correct Answer: B

  https://learn.microsoft.com/en-us/azure/cognitive-services/computer-vision/how-to/use-headpose#detect-head-gestures

  You can detect head gestures like nodding and head shaking by tracking HeadPose changes in real time. You can use this feature as a custom liveness detector.

  Liveness detection is the task of determining that a subject is a real person and not an image or video representation. A head gesture detector could serve as one way to help verify liveness, especially as opposed to an image representation

  of a person.

• Question 200:

  You have an Azure subscription that contains an AI enrichment pipeline in Azure Cognitive Search and an Azure Storage account that has 10 GB of scanned documents and images.

  You need to index the documents and images in the storage account. The solution must minimize how long it takes to build the index.

  What should you do?

  A. From the Azure portal, configure parallel indexing.

  B. From the Azure portal, configure scheduled indexing.

  C. Configure field mappings by using the REST API.

  D. Create a text-based indexer by using the REST API.

  Correct Answer: A

  https://learn.microsoft.com/en-us/azure/search/search-howto-large-index#run-indexers-in-parallel If you partition your data, you can create multiple indexer-data-source combinations that pull from each data source and write to the same search index. Because each indexer is distinct, you can run them at the same time, populating a search index more quickly than if you ran them sequentially.