Courseiva

Microsoft Azure AI Engineer Associate AI-102 (AI-102) — Questions 1–75

988 questions total · 14pages · All types, answers revealed

Take a mock examExam hub

Page 1 of 14

Page 2
1
MCQhard

Refer to the exhibit. You are creating an Azure AI Search index. You want to enable semantic search using the 'content' field as both the title and the content. However, you receive an error that semantic search cannot be configured. What is the most likely reason?

A.The 'content' field is of type 'Edm.String', which is not supported for semantic search.
B.The 'content' field is not searchable.
C.The index does not have a sortable field for the title.
D.The 'contentFields' list is empty; semantic search requires at least one content field or keywords field.
AnswerD

Semantic search requires at least one field in title, content, or keywords; contentFields is empty.

Why this answer

Option D is correct because semantic search in Azure AI Search requires at least one field specified in the 'contentFields' list to provide the textual content for semantic ranking. If the 'contentFields' list is empty, the service cannot extract semantic meaning from the index, resulting in the configuration error. The 'content' field can serve as both title and content if properly assigned, but the error indicates the list itself is missing entries.

Exam trap

The trap here is that candidates may assume the 'content' field is inherently valid for semantic search without realizing the semantic configuration requires explicit assignment of fields to the 'contentFields' list, and an empty list triggers the error regardless of the field's properties.

How to eliminate wrong answers

Option A is wrong because 'Edm.String' is the standard field type for semantic search content; semantic search specifically requires fields of type 'Edm.String' that are also searchable. Option B is wrong because the error message about semantic search configuration does not relate to the field's searchable attribute; a non-searchable field would cause a different error when attempting to use it in queries. Option C is wrong because semantic search does not require a sortable field for the title; the title field only needs to be searchable and of type 'Edm.String', and sorting is irrelevant to semantic ranking.

Full explanation →
2
MCQhard

Refer to the exhibit. You send this request to the Conversational Language Understanding API. The response includes the intent 'BookFlight' with entities 'FromCity: Seattle' and 'ToCity: Boston', but the 'Date' entity is missing. What is the most likely cause?

A.The stringIndexType should be 'Utf16CodeUnit'
B.The API version does not support entity extraction
C.The endpoint is pointing to the wrong deployment
D.The model was not trained to recognize date entities
AnswerD

If the Date entity was not included in training, the model will not extract it.

Why this answer

The utterance includes 'next Tuesday' but the model might not have been trained with date entities or the date entity is not defined in the schema. The API request looks correct. The missing date is not due to a different endpoint or API version.

Full explanation →
3
MCQhard

A company uses Azure AI Language Service with Custom Entity Recognition to extract invoice fields. The model correctly extracts invoice numbers but fails to extract dates in the format 'dd/mm/yyyy'. The training data includes dates in 'mm/dd/yyyy' format. What is the most likely issue?

A.The training data does not contain examples with the 'dd/mm/yyyy' format
B.The dates exceed the maximum entity length
C.The language detection is incorrectly identifying the locale
D.The model is overfitting to invoice numbers
AnswerA

The model learns formats from training data; missing format leads to failure.

Why this answer

The correct answer is C because the model learns the date format from training data; if training data uses a different format, it will not recognize other formats. A (entity length) is not the issue. B (overfitting) would still recognize dates if format matches.

D (language detection) is irrelevant.

Full explanation →
4
MCQeasy

You are designing a knowledge mining solution for customer support emails. The solution must extract the customer's name, issue category, and sentiment from each email. Which two Azure AI services should you combine?

A.Azure AI Bot Service and Azure AI Language
B.Azure AI Document Intelligence and Azure AI Search
C.Azure AI Language and Azure AI Search
D.Azure AI Translator and Azure AI Language
AnswerC

Azure AI Language extracts entities and sentiment; Azure AI Search indexes them.

Why this answer

Azure AI Language provides entity extraction and sentiment analysis. Azure AI Search indexes the extracted data for search. The other services are not directly needed for extraction and indexing.

Full explanation →
5
MCQhard

A hospital uses Azure Cognitive Service for Language to extract medical entities from clinical notes. The extraction accuracy for medication names and dosages is low. The engineer needs to improve performance without adding new training data. Which solution should the engineer implement?

A.Add more training data with annotated entities.
B.Use custom entity recognition with a prebuilt healthcare entity component.
C.Retrain the Text Analytics for Health model with additional labeled data.
D.Increase the confidence threshold for entity extraction.
AnswerB

This combines custom and prebuilt entities to improve accuracy.

Why this answer

Option B is correct because the engineer can use custom entity recognition with a prebuilt healthcare entity component, which leverages the existing Text Analytics for Health model's pre-trained entities (including medication names and dosages) without requiring additional training data. This approach combines the prebuilt healthcare model's high accuracy for medical entities with custom entity recognition to fine-tune extraction for specific clinical notes, improving performance without adding new annotated data.

Exam trap

The trap here is that candidates may assume 'Text Analytics for Health' is a trainable model (like custom NER) and choose Option C, not realizing it is a prebuilt, non-retrainable service that can only be extended via custom entity recognition with a prebuilt component.

How to eliminate wrong answers

Option A is wrong because adding more training data with annotated entities directly contradicts the requirement 'without adding new training data' and would require manual annotation effort. Option C is wrong because retraining the Text Analytics for Health model with additional labeled data is not supported—the Text Analytics for Health model is a prebuilt, non-trainable model that cannot be retrained with custom data; it can only be used as-is or combined with custom entity recognition. Option D is wrong because increasing the confidence threshold for entity extraction would reduce the number of entities returned, potentially missing valid medication names and dosages, and does not improve the underlying model's accuracy—it only filters results more aggressively.

Full explanation →
6
Multi-Selecthard

Which TWO actions can you take to mitigate the risk of generating harmful content when using Azure OpenAI Service? (Choose two.)

Select 2 answers
A.Set a system message that instructs the model to avoid harmful outputs.
B.Fine-tune the model on a dataset of safe examples.
C.Deploy the model in multiple regions.
D.Configure Azure AI Content Safety filters.
E.Increase the maxTokens parameter to allow longer responses.
AnswersA, D

System messages can guide model behavior.

Why this answer

Option A is correct because configuring content filters blocks harmful content. Option D is correct because using system messages to set behavior reduces risk. Option B is wrong as fine-tuning doesn't guarantee safety.

Option C is wrong as increasing maxTokens doesn't mitigate harm. Option E is wrong as multiple deployments don't affect safety.

Full explanation →
7
Multi-Selecteasy

Which TWO capabilities are provided by the Azure AI Language service?

Select 2 answers
A.Text translation.
B.Speech-to-text conversion.
C.Custom text classification.
D.Image captioning.
E.Key phrase extraction.
AnswersC, E

Custom text classification is a feature.

Why this answer

Options B and D are correct. Custom text classification and key phrase extraction are capabilities. Option A is incorrect because Azure AI Language does not do speech recognition.

Option C is incorrect because image captioning is Computer Vision. Option E is incorrect because text translation is Azure AI Translator.

Full explanation →
8
MCQhard

Refer to the exhibit. You deploy this ARM template to create an agent. The agent uses a user-assigned managed identity to call an external weather API. The deployment succeeds but the agent fails to authenticate to the weather API. What is the most likely reason?

A.The resourceId for the managed identity is incorrect.
B.The model provider 'AzureAI' should be 'AzureOpenAI'.
C.The URL parameter is missing the API version.
D.The external weather API is not configured to accept tokens from the managed identity's tenant.
AnswerD

The API must trust the identity's token.

Why this answer

Option D is correct because the weather API needs to trust the managed identity's token, which requires configuring the API as a federated identity credential or using an app registration. Option A is wrong because the resourceId is correctly formatted. Option B is wrong because the model provider is valid.

Option C is wrong because the endpoint is not necessarily incorrect.

Full explanation →
9
MCQeasy

You need to restrict access to an Azure AI Language resource so that only a specific virtual network can call the endpoint. Which configuration should you use?

A.Rotate the shared access keys
B.Enable a service endpoint or private endpoint for the resource
C.Assign a managed identity to the resource
D.Configure an IP firewall rule with the VNet's public IP range
AnswerB

Service endpoints and private endpoints restrict access to specific VNets.

Why this answer

Option B is correct because Azure AI Language resources can be isolated to a specific virtual network by enabling either a service endpoint (via the Microsoft.CognitiveServices service tag) or a private endpoint (using Azure Private Link). This configuration ensures that only traffic originating from the designated VNet can reach the resource's endpoint, effectively blocking all public internet access. Service endpoints provide a direct, optimized route from the VNet to the resource, while private endpoints assign a private IP from the VNet to the resource, making it accessible only within the VNet.

Exam trap

The trap here is that candidates often confuse network-level access controls (service/private endpoints) with authentication mechanisms (keys, managed identities) or IP-based firewalls, mistakenly believing that rotating keys or using managed identities can restrict network access, or that a VNet's public IP range is the same as the VNet's internal address space.

How to eliminate wrong answers

Option A is wrong because rotating shared access keys changes the authentication tokens but does not restrict network-level access; any client with the new keys can still call the endpoint from anywhere on the internet. Option C is wrong because assigning a managed identity enables the resource to authenticate to other Azure services (e.g., Azure Key Vault) without storing credentials, but it does not control which networks can reach the resource's endpoint. Option D is wrong because IP firewall rules with the VNet's public IP range are ineffective for restricting access to a specific virtual network, as VNet traffic typically uses private IPs (RFC 1918) and the public IP of a VNet's NAT gateway or load balancer is not the same as the VNet's internal address space; moreover, IP firewall rules cannot distinguish traffic originating from within the VNet versus other sources using the same public IP range.

Full explanation →
10
MCQmedium

You are building a knowledge mining solution for legal documents stored in Azure Blob Storage. The solution must extract entities, key phrases, and relationships from the documents. Which Azure AI service should you use?

A.Azure AI Document Intelligence
B.Azure AI Translator
C.Azure AI Language
D.Azure AI Search
AnswerC

Azure AI Language includes pre-built capabilities for entity recognition, key phrase extraction, and relationship extraction.

Why this answer

Azure AI Language provides entity extraction, key phrase extraction, and relationship extraction capabilities. Azure AI Document Intelligence is for document analysis and form extraction. Azure AI Search is for indexing and search.

Azure AI Translator is for translation.

Full explanation →
11
MCQeasy

You are testing an Azure OpenAI model with the parameters shown in the exhibit. The model generates very short responses. Which parameter should you modify to allow longer responses?

A.Increase frequency_penalty
B.Increase top_p
C.Increase temperature
D.Increase max_tokens
AnswerD

max_tokens directly controls the maximum length of the response.

Why this answer

The max_tokens parameter controls the maximum number of tokens (words or subwords) the model can generate in a single response. When responses are very short, increasing max_tokens allows the model to produce longer completions up to the specified limit. The other parameters affect randomness, diversity, or probability distribution, not the length cap.

Exam trap

The trap here is that candidates confuse parameters that control output length (max_tokens) with those that control output diversity or creativity (temperature, top_p, frequency_penalty), leading them to incorrectly adjust the latter when the real issue is a token limit.

How to eliminate wrong answers

Option A is wrong because frequency_penalty reduces the likelihood of repeating the same tokens or phrases, which can actually shorten responses by discouraging repetition, not lengthen them. Option B is wrong because top_p (nucleus sampling) controls the cumulative probability threshold for token selection, affecting diversity but not the maximum output length. Option C is wrong because temperature adjusts the randomness of token selection (higher = more creative, lower = more deterministic) and does not impose or remove a length constraint.

Full explanation →
12
MCQhard

You are building a custom named entity recognition (NER) model using Azure AI Language. After labeling 200 documents, you train the model and achieve 85% precision but only 60% recall. Which action is most likely to improve recall?

A.Lower the confidence threshold
B.Increase the training hours
C.Increase the number of labeled documents, especially those containing the target entities
D.Switch to a different Azure AI Language feature
AnswerC

More examples improve recall.

Why this answer

Recall is the ability to find all relevant instances. Adding more labeled examples with the target entities will help the model recognize more patterns. Increasing training hours doesn't guarantee improvement; adjusting confidence threshold lowers precision; using a different service is not necessary.

Full explanation →
13
MCQmedium

You are building a knowledge mining solution for a legal firm to extract clauses from contracts. The contracts are stored as PDFs in Azure Blob Storage. You need to design the solution to minimize cost while ensuring high accuracy for clause extraction. Which approach should you use?

A.Use Azure AI Custom Vision to detect clause regions in scanned documents.
B.Use Azure OpenAI GPT-4 to process each PDF and extract clauses using prompts.
C.Use Azure AI Search with a blob indexer to extract clauses during indexing.
D.Use Azure AI Document Intelligence with a custom extraction model trained on contract clauses.
AnswerD

Document Intelligence provides custom models for key-value pair and table extraction with high accuracy and lower cost than GPT-4.

Why this answer

Option B is correct because Azure AI Document Intelligence (formerly Form Recognizer) offers custom extraction models with high accuracy and is optimized for document understanding. Azure AI Search with blob indexers is for indexing, not extraction. Azure OpenAI GPT-4 can be costlier and may not be as optimized for structured extraction from PDFs.

Custom Vision is for images, not documents.

Full explanation →
14
MCQmedium

Refer to the exhibit. You called the Named Entity Recognition API on a document. Which entity type is "Seattle"?

A.Organization
B.Location
C.Person
D.City
AnswerB

Directly from the exhibit.

Why this answer

The JSON shows "type": "Location" for Seattle.

Full explanation →
15
MCQhard

You are building a generative AI application using Azure OpenAI Service. The application must provide citations for answers retrieved from a set of documents. You need to ensure that each answer includes a reference to the source document. Which configuration should you use?

A.Use Azure OpenAI On Your Data with the 'include citations' option
B.Add a system message requesting citations
C.Implement a custom prompt flow with citation logic
D.Fine-tune the model to include citations
AnswerA

This feature automatically returns source documents as citations.

Why this answer

Option A is correct because Azure OpenAI On Your Data provides a built-in 'include citations' feature that automatically retrieves and appends source document references to the generated answer. This configuration leverages the underlying search index to map each response segment back to the original document, ensuring compliance with citation requirements without custom development.

Exam trap

Microsoft often tests the misconception that a system message or prompt engineering alone can enforce reliable citation behavior, when in fact only a retrieval-augmented generation (RAG) configuration with explicit citation grounding—like Azure OpenAI On Your Data—can guarantee source-linked answers.

How to eliminate wrong answers

Option B is wrong because adding a system message requesting citations only instructs the model to include citations in its output, but the model has no inherent mechanism to reliably retrieve or verify source documents—it may hallucinate citations or omit them entirely. Option C is wrong because implementing a custom prompt flow with citation logic requires significant engineering effort and does not leverage the native, optimized citation pipeline in Azure OpenAI On Your Data, which handles retrieval-augmented generation (RAG) with citation grounding out of the box. Option D is wrong because fine-tuning the model to include citations would require a large, curated training dataset with correct citations and does not guarantee accurate source attribution for new, unseen documents; it also lacks the dynamic retrieval capability needed for document-grounded answers.

Full explanation →
16
MCQmedium

You are designing an AI solution that uses Azure AI Document Intelligence to extract data from invoices. The solution must handle a high volume of documents with varying layouts. Which approach should you use?

A.Use a custom template model with fixed field positions
B.Train a custom neural model with labeled invoices
C.Use the prebuilt invoice model
D.Extract text using OCR and then use regex parsing
AnswerB

Neural models adapt to varying layouts and scale well.

Why this answer

Option B is correct because custom neural models in Azure AI Document Intelligence are designed to handle high volumes of documents with varying layouts. Unlike fixed template models, neural models learn from labeled examples and generalize across different invoice structures, making them ideal for diverse, high-volume scenarios.

Exam trap

The trap here is that candidates often assume the prebuilt invoice model (Option C) is sufficient for all invoice scenarios, but it only works for standard layouts and fails when invoices have custom fields or non-standard structures.

How to eliminate wrong answers

Option A is wrong because custom template models rely on fixed field positions and are brittle when layouts vary; they fail if the invoice format changes even slightly. Option C is wrong because the prebuilt invoice model is limited to standard invoice layouts and cannot adapt to custom or highly variable formats, leading to poor extraction accuracy. Option D is wrong because OCR with regex parsing is a brittle, rule-based approach that cannot handle the semantic understanding required for diverse invoice layouts and fails when fields are not in predictable positions or formats.

Full explanation →
17
MCQhard

You are designing a solution that must extract personally identifiable information (PII) from medical records stored in Azure Blob Storage. The solution must redact the PII before storing the results. Which combination of Azure services should you use?

A.Use Azure AI Language's PII detection feature and a custom Azure Function to redact.
B.Use Azure AI Search with cognitive skills for PII detection.
C.Use Azure OpenAI to detect and redact PII.
D.Use Text Analytics for Health and then manually redact.
AnswerA

PII detection identifies PII, custom function redacts.

Why this answer

Option A is correct because Azure AI Language's PII detection can detect PII, and a custom function can redact it. Option B is incorrect because Azure AI Search does not redact. Option C is incorrect because Text Analytics for Health does not redact.

Option D is incorrect because Azure OpenAI may not have built-in redaction.

Full explanation →
18
MCQmedium

You need to generate a poem using Azure OpenAI. The poem should be about nature and have a cheerful tone. Which parameter should you adjust to influence the tone?

A.system message
B.max_tokens
C.top_p
D.temperature
AnswerA

System message guides the model's overall behavior and tone.

Why this answer

The system message is the correct parameter to influence the tone of the generated poem because it sets the initial context, persona, and behavioral guidelines for the model. By including an instruction like 'You are a cheerful poet who writes about nature in a happy tone,' you directly control the style and emotional quality of the output, which is exactly what the question requires.

Exam trap

Microsoft often tests the distinction between parameters that control randomness (temperature, top_p) versus those that control behavior and style (system message), leading candidates to mistakenly choose temperature as the primary tone influencer.

How to eliminate wrong answers

Option B (max_tokens) is wrong because it controls the maximum length of the generated response, not the tone or style. Option C (top_p) is wrong because it controls nucleus sampling, which affects the diversity of word choices by limiting the cumulative probability of token selection, not the tone. Option D (temperature) is wrong because it controls the randomness of the output (higher values increase creativity, lower values make output more deterministic), but it does not directly set or enforce a specific tone like 'cheerful'.

Full explanation →
19
Multi-Selecthard

Your organization uses Azure AI Vision to analyze surveillance images for security threats. The solution must comply with GDPR. Which TWO actions should you take?

Select 2 answers
A.Enable geo-redundant storage for the Azure AI Vision resource.
B.Enable Customer-Managed Keys (CMK) for encryption.
C.Use private endpoints to access the service.
D.Disable logging of request and response data.
E.Configure the service to use a specific data residency region within the EU.
AnswersD, E

Prevents personal data in logs.

Why this answer

Options B and D are correct because disabling logging ensures personal data is not stored in logs, and using data residency ensures data stays within the EU. Option A is wrong because CMK does not affect logging compliance. Option C is wrong because private endpoints improve security but do not directly address GDPR logging requirements.

Option E is wrong because geo-redundant storage may cause data to leave the EU.

Full explanation →
20
MCQeasy

You need to provide a generative AI solution that can answer questions based on a large set of PDF documents stored in Azure Blob Storage. The solution must support natural language queries and return citations from the documents. Which Azure service combination should you use?

A.Azure Machine Learning and Azure Kubernetes Service
B.Azure Cognitive Search and Azure OpenAI Service with 'on your data'
C.Azure AI Bot Service and Azure Functions
D.Azure AI Document Intelligence and Azure AI Translator
AnswerB

Cognitive Search indexes PDFs; Azure OpenAI uses the index for grounded Q&A with citations.

Why this answer

Option C is correct because Azure Cognitive Search indexes the PDFs and Azure OpenAI on your data uses that index to answer with citations. Option A is wrong because Form Recognizer is for extraction, not Q&A. Option B is wrong because Azure ML is for training, not search.

Option D is wrong because Azure AI Bot Service alone doesn't index documents.

Full explanation →
21
Drag & Dropmedium

Drag and drop the steps to troubleshoot a failed Azure AI Search indexer execution into the correct order.

Drag steps to the numbered slots on the right, or tap a step then tap a slot.

Steps
Order

Why this order

First check history, review errors, verify data source, check skillset, then re-run.

Full explanation →
22
Multi-Selectmedium

You are deploying an Azure AI Language custom text classification model. You need to ensure the model meets performance requirements before promoting it to production. Which two actions should you take? (Choose two.)

Select 2 answers
A.Evaluate the model on a held-out test set that was not used during training.
B.Review the confusion matrix to understand which classes are frequently misclassified.
C.Ensure the model achieves at least 95% accuracy on a cross-validation split.
D.Use the training set to compute accuracy and ensure it is above 90%.
E.Compare the model's performance to a baseline model that always predicts the most common class.
AnswersA, B

A held-out test set gives an unbiased estimate of real-world performance.

Why this answer

Options A and D are correct because evaluating on a held-out test set provides an unbiased performance estimate, and reviewing the confusion matrix helps identify specific misclassifications. Option B is wrong because using the training set for evaluation overestimates performance. Option C is wrong because precision and recall are more informative than just accuracy.

Option E is wrong because comparing to a baseline is good practice but not specific to model evaluation before promotion.

Full explanation →
23
MCQhard

You executed the Azure CLI command shown to create an indexer. However, the indexer fails to run. The error indicates that the data source connection string is invalid. You have verified that the connection string is correct. What is the most likely issue?

A.The --query parameter is incorrectly formatted
B.The skillset name does not exist in the search service
C.The parsingMode should be 'json' for PDF files
D.The data source's connection string is not properly stored or the data source is not configured with the correct credentials
AnswerD

The error indicates invalid connection string; the data source definition likely has a wrong or expired key.

Why this answer

Option D is correct because the Azure CLI command uses single quotes around the parameters JSON, which is correct in Bash, but in PowerShell or other shells, it may cause issues. However, the most common issue is that the command is missing the '--connection-string' parameter or the data source itself has an incorrect connection string. But given the exhibit, the command does not include a '--connection-string' parameter; the data source is referenced by name.

The problem might be that the data source was created with a connection string that is not valid. Option A is wrong because the parsingMode is valid. Option B is wrong because the skillset name is not causing the error.

Option C is wrong because the query parameter is not the issue.

Full explanation →
24
MCQeasy

A company wants to build a solution that can identify and redact personally identifiable information (PII) from customer support transcripts. The solution must handle multiple languages. Which Azure AI service should be used?

A.Azure AI Content Safety
B.Azure AI Document Intelligence
C.Azure AI Translator
D.Azure AI Language - PII Detection
AnswerD

PII Detection identifies and can redact PII in multiple languages.

Why this answer

Option C is correct because Azure AI Language's PII detection and redaction supports multiple languages. Option A is wrong because Azure AI Content Safety is for harmful content. Option B is wrong because Document Intelligence is for forms.

Option D is wrong because Azure AI Translator is for translation.

Full explanation →
25
MCQmedium

A developer is building a custom agent using the Microsoft Bot Framework SDK. The agent must be able to handle multiple turns and maintain context across the conversation. The agent uses dialogs to guide the user through a multi-step process. Which component is responsible for managing the dialog stack and persisting state between turns?

A.The ActivityHandler class that processes incoming activities.
B.The IBot interface implementation.
C.The StateMiddleware component.
D.The DialogSet object that contains dialogs and manages the stack.
AnswerD

DialogSet manages the dialog stack and uses state to persist the stack between turns.

Why this answer

Option C is correct because the DialogSet manages the dialog stack and uses state to persist between turns. Option A is wrong because IBot is the main interface but does not manage dialogs directly. Option B is wrong because StateMiddleware is part of the Bot Framework but not dialog-specific.

Option D is wrong because ActivityHandler handles activities but not dialog management.

Full explanation →
26
Multi-Selecteasy

Which TWO Azure services can be used to detect and redact PII from images containing text? (Choose two.)

Select 2 answers
A.Azure Form Recognizer.
B.Azure Video Analyzer for Media.
C.Azure AI Content Safety.
D.Azure Computer Vision (OCR).
E.Azure Cognitive Search.
AnswersC, D

Can detect and redact sensitive content.

Why this answer

Azure AI Content Safety can detect and redact PII from images containing text by analyzing the extracted text for sensitive information such as personal identifiers and applying redaction. Azure Computer Vision (OCR) provides the underlying text extraction capability, and when combined with Content Safety's PII detection, enables end-to-end redaction of PII from images.

Exam trap

The trap here is that candidates may confuse Azure Form Recognizer's document analysis with general image text processing, overlooking that Content Safety is the specific service for PII detection and redaction, while Computer Vision provides the OCR foundation.

Full explanation →
27
MCQhard

You are deploying a custom image classification model using Azure AI Custom Vision. The model must achieve high accuracy on a dataset with subtle differences between classes. However, the training set is small (200 images per class). Which strategy should you use to improve model performance?

A.Enable hyperparameter tuning and increase the number of iterations
B.Apply aggressive data augmentation and train from scratch
C.Use a larger batch size and train for more epochs
D.Use transfer learning with a pre-trained model and fine-tune
AnswerD

Transfer learning leverages pre-trained weights and is effective with small datasets.

Why this answer

Transfer learning with a pre-trained model is the standard approach for small datasets. Fine-tuning allows the model to leverage learned features.

Full explanation →
28
Multi-Selecthard

You are designing a computer vision solution that must detect and redact personally identifiable information (PII) from images of ID cards. The solution must comply with GDPR. Which THREE components should you include? (Select three.)

Select 3 answers
A.Azure AI Document Intelligence redaction
B.Azure AI Custom Vision
C.Azure AI Vision OCR
D.Azure AI Language PII detection
E.Azure AI Video Indexer
AnswersA, C, D

Can redact detected PII from documents.

Why this answer

Options B, D, and E are correct. Azure AI Vision OCR extracts text from ID cards. Azure AI Language's PII detection identifies PII entities in the extracted text.

Azure AI Document Intelligence's redaction feature (or a custom redaction step) removes the PII. Option A is wrong because Azure AI Video Indexer is for video analysis. Option C is wrong because Azure AI Custom Vision is for image classification/object detection, not text-based PII detection.

Full explanation →
29
MCQhard

You are a data scientist at a healthcare startup. You have deployed a custom object detection model using Azure Custom Vision to detect tumors in MRI scans. The model was trained on 10,000 labeled scans from a single hospital. After deployment, the model performs well on scans from that hospital but poorly on scans from a different hospital with a different MRI machine. The new hospital's scans have slightly different contrast and resolution. The model's precision drops from 0.92 to 0.65, and recall drops from 0.88 to 0.50. You have access to 500 labeled scans from the new hospital. You need to improve the model's performance on the new hospital's data as quickly as possible with minimal effort. What should you do?

A.Collect more labeled scans from the new hospital and train a new model from scratch.
B.Create a new Custom Vision project and train only on the 500 new scans.
C.Apply image preprocessing to normalize the new hospital's scans to match the old hospital's style, then use the existing model.
D.Use the existing model as a starting point and retrain it with the 500 labeled scans from the new hospital.
AnswerD

Transfer learning with new data quickly adapts the model to the new domain with minimal effort.

Why this answer

Option D is correct because Azure Custom Vision supports transfer learning, allowing you to take an existing trained model and retrain it with new labeled data. By using the 500 labeled scans from the new hospital as a training set, you can fine-tune the model to adapt to the different contrast and resolution characteristics without starting from scratch. This approach is the fastest and requires minimal effort, leveraging the previously learned features while incorporating domain-specific adjustments.

Exam trap

The trap here is that candidates may overestimate the need for large datasets or manual preprocessing, failing to recognize that Azure Custom Vision's built-in transfer learning is designed to efficiently adapt models with minimal new data.

How to eliminate wrong answers

Option A is wrong because collecting more labeled scans and training a new model from scratch is time-consuming and resource-intensive, not the quickest or minimal-effort solution. Option B is wrong because creating a new Custom Vision project and training only on 500 scans ignores the valuable knowledge from the original 10,000 scans, leading to a model with insufficient data and likely poor generalization. Option C is wrong because applying image preprocessing to normalize the new hospital's scans to match the old hospital's style is a manual, error-prone process that may not fully address the underlying domain shift and does not leverage the labeled data for supervised adaptation.

Full explanation →
30
MCQhard

Your team is developing an AI-powered document summarization solution using Azure OpenAI. You need to ensure that the solution complies with Microsoft's Responsible AI principles, specifically transparency. Which configuration should you implement?

A.Configure diagnostic logging to capture all model inputs and outputs.
B.Fine-tune the model on a custom dataset to improve accuracy.
C.Enable content filtering with severity levels high and medium.
D.Add a system message that informs users the summary is generated by AI.
AnswerD

Transparency requires clear disclosure of AI involvement.

Why this answer

Transparency under Microsoft's Responsible AI principles requires that users are aware when they are interacting with an AI system. Adding a system message that explicitly states the summary is AI-generated fulfills this disclosure requirement. Diagnostic logging (A) aids in accountability and debugging but does not directly inform the user.

Fine-tuning (B) improves accuracy but does not address transparency. Content filtering (C) mitigates harmful outputs but does not disclose AI involvement.

Exam trap

The trap here is that candidates confuse 'transparency' with 'accountability' or 'safety' and select diagnostic logging or content filtering, not realizing that transparency specifically requires user-facing disclosure of AI involvement.

How to eliminate wrong answers

Option A is wrong because diagnostic logging captures inputs and outputs for auditing and debugging, but it does not communicate to the end user that the content is AI-generated, which is the core requirement of transparency. Option B is wrong because fine-tuning the model on a custom dataset enhances performance and relevance but has no role in informing users about AI authorship. Option C is wrong because enabling content filtering with severity levels high and medium is a safety measure to block harmful content, not a mechanism for disclosing AI involvement to users.

Full explanation →
31
Multi-Selectmedium

A company is building an agent that uses Azure OpenAI to answer questions from a large document library. The agent must use a Retrieval Augmented Generation (RAG) pattern. Which TWO actions should the team take to implement RAG effectively?

Select 2 answers
A.Ensure the model is large enough to memorize the entire document library.
B.Fine-tune the Azure OpenAI model on the document library.
C.Index the documents into a vector database like Azure Cognitive Search.
D.Train a custom language model from scratch.
E.Use a retrieval step to fetch relevant document chunks before generating a response.
AnswersC, E

Indexing enables efficient retrieval of relevant content.

Why this answer

Option C is correct because indexing documents into a vector database like Azure Cognitive Search enables efficient similarity search over embeddings, which is the retrieval foundation of RAG. This allows the system to quickly find the most relevant document chunks based on semantic meaning, rather than relying on the model to memorize or be fine-tuned on the entire library.

Exam trap

The trap here is that candidates often confuse fine-tuning (which adapts model behavior) with RAG (which augments prompts with retrieved data), leading them to select Option B instead of understanding that RAG requires an external retrieval step and vector index.

Full explanation →
32
MCQhard

You are designing a solution that uses Azure AI Vision to extract text from scanned invoices. The invoices vary in layout and include both printed and handwritten fields. The solution must achieve high accuracy with minimal manual labeling. Which approach should you recommend?

A.Use the Read API to extract all text and then use a custom regex to parse fields.
B.Use the Azure AI Document Intelligence prebuilt invoice model.
C.Train a Custom Vision object detection model to locate fields.
D.Label hundreds of invoices and train a custom Azure AI Document Intelligence model.
AnswerB

Prebuilt invoice model handles varying layouts and mixed text.

Why this answer

Option D is correct because Azure AI Document Intelligence (formerly Form Recognizer) is designed for extracting fields from documents with varying layouts and supports both printed and handwritten text. Option A is wrong because the Read API extracts all text but does not extract specific fields. Option B is wrong because Custom Vision is for object detection, not document analysis.

Option C is wrong because it requires significant manual labeling.

Full explanation →
33
Multi-Selecthard

Which THREE factors should you consider when choosing between Azure AI Custom Vision and Azure AI Vision pre-built models for an image classification task? (Choose three.)

Select 3 answers
A.Availability of labeled training data specific to the domain
B.Image format support (JPEG, PNG)
C.Whether the required labels are covered by the pre-built model
D.Need for real-time inference latency
E.Need to iterate and retrain the model over time
AnswersA, C, E

Custom Vision requires labeled data; pre-built models do not.

Why this answer

Custom Vision is suitable when you have custom labels not covered by pre-built models, and when you have sufficient training data. Pre-built models are better when you need immediate deployment without data. Option B is wrong because real-time requirements can be met by both services.

Option D is wrong because both services support common image formats.

Full explanation →
34
MCQhard

You see the exhibit representing an Azure Bot resource configuration. The bot is not responding to user messages. What should you verify first?

A.Check that the endpoint URL is correct and the web app is running
B.Ensure that LUIS app IDs are provided
C.Verify that the Application Insights key is valid
D.Confirm that the msaAppId is a valid GUID
AnswerA

The bot's web app must be running and the endpoint must be reachable.

Why this answer

The most common reason a bot fails to respond to user messages is that the endpoint URL configured in the Azure Bot resource does not match the actual URL of the running web app, or the web app itself is stopped or unresponsive. Without a correct and reachable endpoint, the Bot Framework Service cannot forward messages to the bot's message handler, causing a complete communication breakdown.

Exam trap

The trap here is that candidates often jump to authentication or AI service configuration issues, but the most immediate and common cause of a non-responsive bot is a simple connectivity or endpoint misconfiguration.

How to eliminate wrong answers

Option B is wrong because LUIS app IDs are only required if the bot uses Language Understanding for intent detection; they are not necessary for basic message handling. Option C is wrong because Application Insights is used for telemetry and monitoring, not for core message routing; an invalid key would not prevent the bot from responding. Option D is wrong because the msaAppId (Microsoft App ID) is used for authentication with the Bot Framework Service, but if it were invalid, the bot would typically fail to register or authenticate, not silently ignore messages; the endpoint and web app availability are more fundamental.

Full explanation →
35
MCQmedium

You are a lead AI engineer for a global retail company. The company is building an AI-powered customer support chatbot using Microsoft Foundry. The chatbot must answer product questions, process returns, and escalate to human agents when needed. The solution uses Azure AI Language for intent recognition and Azure AI Bot Service for bot orchestration. During testing, the chatbot fails to understand customer queries about return policies. The intents 'ProductInquiry' and 'ReturnRequest' are defined, but the model often confuses them. You need to improve intent classification accuracy. The development team has already collected 500 sample utterances for each intent. You have a budget to collect additional data. What should you do?

A.Use Azure AI Translator to translate utterances into multiple languages
B.Collect additional utterances that are similar to the confusing ones and retrain
C.Increase the confidence threshold in the bot configuration
D.Create a new custom language project and retrain from scratch
AnswerB

Adds more examples to differentiate intents.

Why this answer

Option B is correct because collecting additional utterances that are similar to the confusing ones directly addresses the ambiguity between the 'ProductInquiry' and 'ReturnRequest' intents. By providing more representative examples of edge cases where the intents overlap, the Azure AI Language model can better learn the subtle linguistic patterns that distinguish them, thereby improving classification accuracy without requiring a complete retraining from scratch.

Exam trap

The trap here is that candidates often assume increasing the confidence threshold (Option C) will fix misclassifications, but this only adjusts the prediction cutoff and does not improve the underlying model's ability to differentiate intents, which is a data quality issue, not a threshold tuning issue.

How to eliminate wrong answers

Option A is wrong because translating utterances into multiple languages does not resolve confusion between two intents in the same language; it only adds multilingual support, which is irrelevant to the core issue of intent ambiguity. Option C is wrong because increasing the confidence threshold would only reject more low-confidence predictions, not improve the model's ability to distinguish between similar intents; it could actually increase false negatives by requiring higher confidence for correct classifications. Option D is wrong because creating a new custom language project and retraining from scratch is unnecessary and wasteful; the existing project already has 500 utterances per intent, and the problem is specifically about confusing utterances, not a fundamental flaw in the project setup.

Full explanation →
36
MCQeasy

You are building a chatbot using Microsoft Copilot Studio that needs to answer questions based on content from a set of technical manuals stored as PDFs. The content must be indexed and made available to the chatbot. You need to configure the knowledge source for the chatbot. What should you do?

A.Upload the PDFs to a SharePoint library and use SharePoint as the knowledge source.
B.Create an Azure AI Search index from the PDFs and connect it to Copilot Studio as a knowledge source.
C.Add a web search connector to the chatbot to search for information online.
D.Store the content in Azure SQL Database and use it as a knowledge source.
AnswerB

Azure AI Search indexes content and can be used as a knowledge source.

Why this answer

Microsoft Copilot Studio can use Azure AI Search as a knowledge source to index and retrieve content from PDFs. Option A is wrong because the Web search is for public information. Option B is wrong because SharePoint is not the direct indexer for PDFs.

Option D is wrong because Azure SQL Database is not for document content.

Full explanation →
37
MCQhard

You are developing a solution that uses Azure AI Video Indexer to analyze surveillance videos for suspicious activity. The solution must generate alerts when a person is detected in a restricted area. Which feature should you use?

A.Azure AI Video Indexer sentiment analysis
B.Azure AI Video Indexer people detection and tracking
C.Azure AI Face identify API
D.Object detection in Azure AI Vision
AnswerB

Tracks people movements and can trigger alerts.

Why this answer

Option B is correct because Azure AI Video Indexer's detection of people and their movements can be used to trigger alerts when a person is in a restricted area. Option A is wrong because object detection alone does not track location over time. Option C is wrong because face identification is for identifying specific individuals, not for location-based alerts.

Option D is wrong because sentiment analysis is for emotions, not location.

Full explanation →
38
MCQeasy

You are using Azure OpenAI Service to generate product descriptions. The output is often too verbose. You need to reduce the length of generated text without changing the model. Which parameter should you adjust?

A.Max tokens
B.Frequency penalty
C.Temperature
D.Top-p (nucleus sampling)
AnswerA

Max tokens limits the length of the generated response.

Why this answer

Max tokens controls the total length of the generated response by capping the number of tokens (words/subwords) the model can output. Reducing this value directly truncates the output, making descriptions shorter without altering the model or its behavior. Other parameters influence randomness or repetition but do not enforce a strict length limit.

Exam trap

The trap here is that candidates confuse parameters that affect output style (temperature, top-p, frequency penalty) with the one parameter that directly controls output length (max tokens), leading them to choose a parameter that changes how the model writes rather than how much it writes.

How to eliminate wrong answers

Option B (Frequency penalty) is wrong because it reduces the likelihood of repeating the same tokens or phrases, which can change content style but does not enforce a maximum output length. Option C (Temperature) is wrong because it controls the randomness of token selection (higher = more creative, lower = more deterministic), not the number of tokens generated. Option D (Top-p) is wrong because it limits the cumulative probability of token choices (nucleus sampling), affecting diversity but not the total token count.

Full explanation →
39
MCQmedium

Your company has a large repository of scanned invoices in PDF format. You need to extract invoice number, date, total amount, and vendor name from these PDFs. Which Azure AI service should you use?

A.Azure AI Language
B.Azure AI Document Intelligence
C.Azure AI Search
D.Azure AI Vision
AnswerB

Document Intelligence has a pre-built invoice model that extracts invoice-specific fields.

Why this answer

Azure AI Document Intelligence (formerly Form Recognizer) is designed for extracting structured data from forms and documents. Pre-built invoice model can extract invoice fields. Azure AI Language is for text analytics on extracted text.

Full explanation →
40
Multi-Selectmedium

Which TWO actions should you take to ensure compliance with data residency requirements when using Azure AI services across multiple regions?

Select 2 answers
A.Use a global Azure AI service resource and configure geo-replication
B.Configure the AI service's data residency settings to restrict data to the region
C.Enable Azure Front Door to route traffic to the nearest region
D.Disable data replication for the AI service
E.Deploy separate instances of the AI service in each required region
AnswersB, E

Some services offer data residency settings to limit data movement.

Why this answer

Option A is correct because deploying resources in the required region ensures data stays there. Option E is correct because configuring data residency settings in the service prevents cross-region data movement. Option B is incorrect because data may still move across regions.

Option C is incorrect because Azure Front Door does not affect data residency. Option D is incorrect because disabling redundancy does not guarantee data residency.

Full explanation →
41
MCQmedium

Your company runs a global e-commerce platform. You are building a chatbot using Azure AI Language's conversational language understanding (CLU) to handle customer requests in multiple languages. The bot must support English, German, and Japanese. You have labeled training data in English only. The deadline is tight, and you want to minimize manual labeling. You also need to ensure that the bot can gracefully handle unsupported languages (e.g., French) by directing the user to a human agent. You have access to Azure AI Translator. Which approach should you take?

A.Use a single CLU project with English data only. Translate the English training data into German and Japanese using Azure AI Translator, then train a single multilingual model by including the translated data.
B.Use a single CLU project with English data only. Before calling CLU, translate non-English user input to English using Azure AI Translator. For unsupported languages, detect language and route to human agent.
C.Use a single CLU project with multilingual option enabled, train on English data only. Configure the bot to detect the language of user input; if it is English, German, or Japanese, route to CLU; otherwise, route to a human agent.
D.Build separate CLU projects for English, German, and Japanese. Label training data in each language by translating the English data using Azure AI Translator.
AnswerC

The multilingual option allows the model to predict intents in English, German, and Japanese without additional labeled data. Language detection ensures unsupported languages are handled appropriately.

Why this answer

Option D is correct because it uses the multilingual CLU model for English, German, and Japanese (the model can predict in those languages without additional labeled data) and uses a language detection step to route unsupported languages to a human agent. Option A is wrong because building separate projects for each language requires labeling in each language, which is not minimal effort. Option B is wrong because translating user input introduces latency and potential translation errors.

Option C is wrong because translating the English data and training separate models increases effort and may not work well due to translation quality.

Full explanation →
42
MCQmedium

You are developing a solution for a hospital that uses the Face API to identify patients from photos taken at check-in. The system must be HIPAA compliant. You need to ensure that face data is protected and not retained longer than necessary. The hospital wants to use the Face API for identification only during the patient's visit. After discharge, the face data should be deleted. What is the recommended approach?

A.Use the Face API with a subscription key and export face data to local storage, then delete from the cloud.
B.Store the face IDs in a database and delete them manually after discharge.
C.Create a Person Group for each day, add patients, and delete the Person Group after the day ends.
D.Use the Face API Identify operation and then delete the face ID from the Person Group.
AnswerC

Person Groups can be deleted to remove all face data at once.

Why this answer

Option A is correct because the Face API supports person group management; you can create a temporary person group for each visit, then delete it after discharge. Option B is wrong because storing face IDs indefinitely is not compliant. Option C is wrong because exporting to local storage and deleting may lose traceability.

Option D is wrong because the Face API does not allow face ID deletion without deleting the person group.

Full explanation →
43
Multi-Selectmedium

A healthcare organization is deploying a solution using Azure AI Language to extract medical entities from clinical notes. The solution must comply with HIPAA and support the following requirements: extract medication names, dosages, and frequencies; identify patient conditions; and recognize negated terms (e.g., 'no sign of infection'). Which THREE Azure AI Language features should the organization use?

Select 3 answers
A.PII detection
B.Prebuilt NER for Healthcare
C.Prebuilt NER for Finance
D.Negation detection
E.Custom Named Entity Recognition (NER)
AnswersB, D, E

Prebuilt NER for Healthcare recognizes common clinical entities such as conditions, symptoms, and procedures.

Why this answer

The correct features are: Custom Named Entity Recognition (NER) to extract custom medical entities like medication names, dosages, and frequencies; Prebuilt NER for Healthcare to identify patient conditions and other clinical entities; and Negation detection to identify negated terms. The other options are not relevant: Prebuilt NER for Finance handles financial entities; PII detection extracts personal information, not clinical entities.

Full explanation →
44
MCQhard

You are deploying an Azure AI Search solution for a global e-commerce platform. The index must support real-time updates from a product catalog stored in Azure Cosmos DB, with a maximum indexing latency of 10 seconds. The solution must also handle 5000 queries per second (QPS) during peak hours. What should you configure?

A.Use the Storage Optimized L2 tier with 6 partitions and 4 replicas.
B.Use the Free tier and enable change tracking on Cosmos DB.
C.Use Standard S3 tier with 12 replicas and enable incremental indexing.
D.Use Standard S2 tier with 3 replicas and indexer scheduling every 5 minutes.
AnswerC

S3 supports high throughput; replicas multiply QPS; incremental indexing reduces latency.

Why this answer

Option C is correct because the Standard S3 tier supports high query volumes (up to 5000 QPS) with 12 replicas, and incremental indexing enables near-real-time updates from Cosmos DB by processing only changed documents, keeping latency under 10 seconds. The combination of sufficient replicas for throughput and incremental indexing for low-latency indexing meets both requirements.

Exam trap

The trap here is that candidates may confuse 'incremental indexing' with 'indexer scheduling,' assuming a scheduled indexer can meet low-latency requirements, but scheduling introduces fixed intervals (e.g., 5 minutes) that violate the 10-second latency constraint.

How to eliminate wrong answers

Option A is wrong because the Storage Optimized L2 tier is designed for large storage capacity with lower query throughput, not for high QPS (5000) or low-latency indexing; 6 partitions and 4 replicas cannot sustain 5000 QPS. Option B is wrong because the Free tier is limited to 1 partition, 1 replica, and 3 indexes, with a maximum of 10 QPS, making it incapable of handling 5000 QPS or real-time updates. Option D is wrong because the Standard S2 tier with 3 replicas supports only up to 1500 QPS (500 per replica), and indexer scheduling every 5 minutes introduces a minimum 5-minute latency, far exceeding the 10-second requirement.

Full explanation →
45
Drag & Dropmedium

Drag and drop the steps to configure a multi-region disaster recovery for Azure Cognitive Services into the correct order.

Drag steps to the numbered slots on the right, or tap a step then tap a slot.

Steps
Order

Why this order

Start by creating resources in both regions, then set up replication, configure failover routing, monitor health, and test.

Full explanation →
46
MCQhard

Your team is developing a chatbot using Azure AI Bot Service with Language Understanding (LUIS). You need to improve intent recognition for user queries that contain typos. What should you recommend?

A.Add synonym lists for common misspellings
B.Use Azure Cognitive Search to correct typos
C.Enable Bing Spell Check in the LUIS app settings
D.Train the model with many examples containing typos
AnswerC

Corrects typos automatically before processing.

Why this answer

Option C is correct because enabling Bing Spell Check in the LUIS app settings allows the service to automatically correct typos in user utterances before they are processed for intent recognition. This built-in feature integrates directly with LUIS at the endpoint, correcting misspelled words to improve the accuracy of intent and entity extraction without requiring manual data augmentation or external services.

Exam trap

The trap here is that candidates often assume training with typos (Option D) is the best way to handle misspellings, but Microsoft explicitly recommends using Bing Spell Check as a separate preprocessing layer rather than bloating the training data with error-prone examples.

How to eliminate wrong answers

Option A is wrong because synonym lists in LUIS are used to map different words that have the same meaning (e.g., 'car' and 'automobile'), not to handle typos or misspellings; they do not correct character-level errors. Option B is wrong because Azure Cognitive Search is a search indexing and query service, not a spell-checking tool; it cannot be used to correct typos in real-time LUIS utterances. Option D is wrong because while training with many examples containing typos might help the model learn some patterns, it is inefficient and does not generalize well to unseen typos; LUIS's built-in Bing Spell Check is the recommended and more robust approach.

Full explanation →
47
MCQeasy

A developer wants to use Azure OpenAI to generate text from a prompt. Which parameter controls the diversity of the generated output?

A.presence_penalty
B.frequency_penalty
C.temperature
D.max_tokens
AnswerC

Temperature controls randomness and diversity.

Why this answer

Temperature is the parameter that directly controls the randomness or diversity of the generated output by scaling the logits before applying the softmax function. A higher temperature (e.g., 1.0) increases the probability of less likely tokens, producing more creative and varied responses, while a lower temperature (e.g., 0.1) makes the output more deterministic and focused.

Exam trap

The trap here is that candidates often confuse frequency_penalty or presence_penalty with controlling diversity, but those parameters address repetition and topic novelty, not the fundamental randomness of token selection, which is exclusively governed by temperature.

How to eliminate wrong answers

Option A is wrong because presence_penalty penalizes tokens that have already appeared in the text so far, encouraging the model to introduce new topics or avoid repetition, but it does not directly control the overall diversity or randomness of the token selection. Option B is wrong because frequency_penalty reduces the likelihood of tokens based on how frequently they have occurred in the generated text, which helps prevent repetitive phrasing but does not adjust the probability distribution's entropy like temperature does. Option D is wrong because max_tokens simply limits the maximum number of tokens in the generated response and has no effect on the diversity or randomness of the output.

Full explanation →
48
Multi-Selectmedium

Which THREE are valid parameters when calling the Azure OpenAI Service chat completions API?

Select 3 answers
A.messages
B.index_name
C.temperature
D.max_tokens
E.embedding_model
AnswersA, C, D

Required input.

Why this answer

The `messages` parameter is required in the Azure OpenAI Service chat completions API call. It defines the conversation history and user input as an array of message objects, each with a `role` (system, user, assistant) and `content`. Without this parameter, the API cannot determine the context or prompt for generating a response.

Exam trap

Microsoft often tests the distinction between parameters for different Azure OpenAI API endpoints (chat completions vs. embeddings vs. search), so candidates may confuse `index_name` or `embedding_model` as valid chat completions parameters due to their familiarity with other Azure AI services.

Full explanation →
49
MCQhard

A retail company uses Azure Computer Vision to analyze in-store camera feeds to count customers. The solution uses the Detect API on individual frames. Recently, the counts have been inaccurate due to overlapping people. Which action should you take to improve accuracy?

A.Train a Custom Vision object detection model with images of crowded scenes.
B.Use the Analyze API instead of Detect API.
C.Increase the confidence threshold in the Detect API call.
D.Switch to Azure Video Indexer for people counting.
AnswerA

Custom model can learn to detect partially occluded people.

Why this answer

Option B is correct because Azure Custom Vision can be trained with images of crowded scenes to better detect overlapping people. Option A is wrong because increasing the confidence threshold may reduce false positives but doesn't handle overlap. Option C is wrong because Video Indexer is for video insights, not real-time counting.

Option D is wrong because the Detect API is already being used; the issue is with the pre-built model's limitations.

Full explanation →
50
MCQhard

A financial services company uses Azure AI Language's custom text classification to categorize loan applications as 'Approved', 'Denied', or 'Review Required'. The model is trained on historical data but is producing poor accuracy on new applications. The data scientist suspects data leakage between training and test sets. What should the data scientist do to validate this?

A.Increase the training dataset size and retrain the model.
B.Use k-fold cross-validation during training.
C.Adjust the classification confidence threshold.
D.Split the data chronologically and ensure no overlapping data between train and test sets.
AnswerD

Chronological split prevents future data from leaking into training.

Why this answer

Option B is correct because evaluating the model with a holdout set that was never used in training is the standard way to detect data leakage. Option A is wrong because more training data won't fix leakage. Option C is wrong because cross-validation can still have leakage if leakage is in the data splitting.

Option D is wrong because adjusting confidence thresholds doesn't fix leakage.

Full explanation →
51
Multi-Selectmedium

You are designing an Azure Cognitive Search solution that indexes customer support tickets. The index must include a field for 'sentiment' that is populated from an AI enrichment pipeline. Which TWO actions are required to achieve this?

Select 2 answers
A.Add a built-in Sentiment skill to the skillset.
B.Implement a custom skill to normalize sentiment values.
C.Create a custom sentiment analysis skill using Azure AI Language.
D.Define a 'sentiment' field in the index with type Collection(Edm.String).
E.Configure an output field mapping in the indexer to map the sentiment output to the index field.
AnswersA, E

The sentiment skill generates sentiment scores.

Why this answer

Options A and C are correct. The enrichment pipeline must include a built-in sentiment skill (A), and a field mapping must be added to map the sentiment output to the index field (C). Option B is wrong because the sentiment skill is already built-in.

Option D is wrong because a custom skill is not required. Option E is wrong because the index definition defines the field but does not populate it.

Full explanation →
52
MCQeasy

A company uses this skillset in an Azure AI Search enrichment pipeline. They notice that the enrichment pipeline fails when processing a document larger than 5000 characters. What is the most likely cause?

A.The maximum page length is too small
B.The default language code is not supported
C.The text split mode should be 'sentences'
D.The output field mapping is missing or incorrect
AnswerD

The output 'pages' must be mapped to a collection field in the index.

Why this answer

The enrichment pipeline fails because the output field mapping is missing or incorrect. When a skillset processes documents, the output of each skill must be explicitly mapped to an index field; if this mapping is absent or misconfigured, the pipeline cannot store the enriched data and fails, especially for larger documents that produce more output data.

Exam trap

The trap here is that candidates often attribute pipeline failures to text splitting or language settings, but the real issue is the missing output field mapping, which is a common misconfiguration in skillset definitions.

How to eliminate wrong answers

Option A is wrong because the maximum page length setting in the text split skill controls chunk size, not the overall document size limit; a 5000-character document is well within typical limits. Option B is wrong because an unsupported language code would cause a language detection or translation skill error, not a generic pipeline failure tied to document size. Option C is wrong because the text split mode (e.g., 'pages' vs 'sentences') affects how text is chunked, but does not cause a pipeline failure solely due to document size; the failure is related to output mapping, not splitting logic.

Full explanation →
53
Multi-Selectmedium

You are deploying a knowledge mining solution using Azure AI Search and Azure AI Document Intelligence. The solution must extract text from scanned documents, identify named entities, and index the content. You need to configure the skillset. Which TWO built-in skills should you include in the skillset?

Select 2 answers
A.Merge skill
B.LanguageDetection skill
C.OCR skill
D.EntityRecognition skill
E.KeyPhraseExtraction skill
AnswersC, D

Extracts text from scanned documents.

Why this answer

The OCR skill extracts text from scanned images. The EntityRecognition skill identifies named entities. The Merge skill is not required because OCR output is already text.

The KeyPhraseExtraction skill extracts key phrases, not entities. The LanguageDetection skill detects language, not entities.

Full explanation →
54
Multi-Selecthard

You are using Microsoft Purview to create a knowledge map of your organization's data assets. The solution must automatically scan and classify sensitive data in Azure Blob Storage. You need to configure the scanning and classification. Which THREE actions should you perform?

Select 3 answers
A.Run a full scan of the Blob Storage to discover and classify data.
B.Create custom classification rules for sensitive data types.
C.Apply sensitivity labels to the classified data.
D.Create a scan rule set that includes the desired classification rules.
E.Register the Azure Blob Storage account as a data source in Purview.
AnswersA, D, E

Scanning is required to apply classifiers.

Why this answer

Register a data source (Blob Storage) in Purview. Create a scan rule set for classification. Run a full scan to classify data.

Option A is wrong because custom classification rules are not required; built-in classifiers can be used. Option D is wrong because sensitivity labels are set after classification, but not a direct action for scanning. Option E is wrong because configuring a firewall is a prerequisite but not part of classification scanning.

Full explanation →
55
MCQhard

You are implementing a knowledge mining solution with Azure AI Search that ingests data from Azure Blob Storage. The pipeline includes a custom skill that calls an external API for specialized entity extraction. The custom skill sometimes returns HTTP 429 (Too Many Requests). How should you handle this to ensure reliable indexing?

A.Reduce the batch size in the indexer
B.Increase the skill timeout
C.Configure a retry policy on the custom skill
D.Schedule the indexer to run less frequently
AnswerC

A retry policy with exponential backoff handles 429 errors by retrying after delays.

Why this answer

Option D is correct because the indexer retry policy can be configured to handle transient failures, including 429 errors, by retrying after a delay. Option A is incorrect because reducing the batch size might not help if the API rate limit is per request. Option B is incorrect because increasing timeouts does not address rate limiting.

Option C is incorrect because scheduling the indexer less frequently does not prevent failures during execution.

Full explanation →
56
MCQmedium

You are designing a knowledge mining solution for a legal firm. The solution must index contracts and support queries like 'find all contracts where the governing law is New York'. The contracts are stored as PDFs in Azure Blob Storage. Which approach should you use?

A.Use Azure AI Search with a custom skill to extract the governing law field
B.Use Azure AI Search with vector search
C.Use Azure AI Search with a built-in skillset for OCR
D.Use Azure AI Search with full-text search on the extracted text
AnswerA

Custom skills can extract specific fields for indexing.

Why this answer

Option C is correct because Azure AI Search with a custom skill to extract the governing law field allows indexing that specific metadata. Option A is wrong because full-text search on raw text would not distinguish the governing law field. Option B is wrong because skillsets can extract fields, not just enrich.

Option D is wrong because vector search is for semantic similarity, not exact field extraction.

Full explanation →
57
MCQmedium

A retail company uses Azure Computer Vision to analyze customer traffic in stores. They process images from security cameras using the OCR API to detect product labels. Recently, the OCR accuracy has decreased for images with poor lighting. Which pre-processing step should the company implement to improve OCR accuracy?

A.Convert images to grayscale before sending to OCR API.
B.Increase the image resolution before calling OCR API.
C.Adjust brightness and contrast of images using image processing.
D.Reduce image size to decrease noise.
AnswerC

Improves visibility of text in low-light conditions, enhancing OCR accuracy.

Why this answer

Option C is correct because poor lighting directly reduces the contrast between text and background, which is critical for OCR accuracy. Adjusting brightness and contrast improves the signal-to-noise ratio of the text regions, making character edges more distinct for the Azure Computer Vision OCR engine. This pre-processing step compensates for the lighting deficiency without altering the fundamental image content that the API relies on.

Exam trap

The trap here is that candidates confuse image quality improvements (like resolution or noise reduction) with the specific need to correct lighting-induced contrast loss, which is a distinct pre-processing requirement for OCR in poor illumination.

How to eliminate wrong answers

Option A is wrong because converting to grayscale removes color information that can help distinguish text from similarly-toned backgrounds, and it does not address the root cause of low contrast due to poor lighting. Option B is wrong because increasing resolution does not fix the underlying contrast problem; it may even amplify noise and increase API processing time without improving text legibility. Option D is wrong because reducing image size discards pixel detail, which can make small or thin text characters unreadable for OCR, and it does not mitigate the effects of poor lighting.

Full explanation →
58
MCQhard

Refer to the exhibit. You deployed a Conversational Language Understanding (CLU) project. A user says 'hi, where is my order?' The model returns a single intent 'Greeting' with confidence 0.85. You need the model to detect both intents. What should you change?

A.Lower the confidenceThreshold to 0.5.
B.Add more utterances to the 'OrderStatus' intent.
C.Set 'multipleIntents' to true.
D.Change the project kind to 'Orchestration'.
AnswerC

Enabling multipleIntents allows the model to return multiple intents.

Why this answer

Option C is correct because the Conversational Language Understanding (CLU) service supports multi-intent detection only when the 'multipleIntents' flag is explicitly set to true in the project settings. By default, CLU returns only the highest-confidence intent; enabling this flag allows the model to return all intents whose confidence exceeds the threshold, enabling detection of both 'Greeting' and 'OrderStatus' from a single utterance.

Exam trap

The trap here is that candidates often confuse the confidenceThreshold parameter with the mechanism for returning multiple intents, assuming lowering the threshold will surface additional intents, when in fact the multipleIntents flag must be enabled first.

How to eliminate wrong answers

Option A is wrong because lowering the confidenceThreshold to 0.5 would still only return a single intent (the highest-confidence one) unless multipleIntents is enabled; the threshold controls the minimum confidence for returned intents but does not enable multi-intent detection. Option B is wrong because adding more utterances to the 'OrderStatus' intent improves its training data but does not change the inference behavior to return multiple intents; the model will still only output the top-scoring intent. Option D is wrong because changing the project kind to 'Orchestration' is used to route utterances to different language services (e.g., LUIS, QnA Maker) rather than enabling multi-intent detection within a single CLU project; it addresses a different architectural need.

Full explanation →
59
Multi-Selecthard

An agent uses Azure OpenAI with function calling to perform actions. The agent is not executing functions correctly. Which THREE factors should the team check to diagnose the issue?

Select 3 answers
A.The temperature parameter is set too high.
B.The token limit is too low, truncating the function definitions.
C.The function parameter schemas are incorrect or incomplete.
D.The function descriptions are ambiguous or missing.
E.The model version is outdated.
AnswersB, C, D

Low token limits can cut off function definitions.

Why this answer

Option B is correct because if the token limit is too low, the model may not receive the full function definitions, causing it to omit or misinterpret available functions. This truncation prevents the model from correctly selecting or formatting function calls, leading to execution failures.

Exam trap

Microsoft often tests the misconception that temperature or model version are primary causes for function-calling failures, when in reality the core issues are token limits, schema correctness, and description clarity.

Full explanation →
60
MCQeasy

You need to translate a document from English to Spanish using Azure OpenAI. Which parameter should you include in the prompt to specify the target language?

A.temperature
B.system message
C.max_tokens
D.user message
AnswerD

User message should contain the translation instruction.

Why this answer

Option D is correct because in Azure OpenAI, the user message is where you provide the instruction or context for the model, including specifying the target language for translation. By including 'Translate the following English text to Spanish' in the user message, you direct the model to perform the language translation task. The other parameters control generation behavior (temperature, max_tokens) or set the model's persona (system message), but do not specify the target language.

Exam trap

The trap here is that candidates confuse the system message (which sets overall behavior) with the user message (which carries the specific instruction), leading them to incorrectly select system message as the parameter for specifying the target language.

How to eliminate wrong answers

Option A is wrong because temperature controls the randomness of the model's output, not the target language. Option B is wrong because the system message sets the assistant's behavior or persona (e.g., 'You are a helpful translator'), but does not specify the target language for a specific translation request. Option C is wrong because max_tokens limits the length of the generated response, not the language of the output.

Full explanation →
61
MCQmedium

You are deploying a question answering solution using Azure AI Language. The solution must be able to provide answers from a set of frequently asked questions (FAQs) in PDF format. What should you do?

A.Use Azure AI Search with cognitive skills.
B.Create a custom question answering project and add the PDF as a source.
C.Use Azure OpenAI with a system prompt containing the PDFs.
D.Use the pre-built question answering in Azure AI Language.
AnswerB

Custom question answering can ingest PDFs.

Why this answer

Option C is correct because custom question answering can ingest PDFs and extract Q&A pairs. Option A is incorrect because Azure AI Search is for search, not question answering. Option B is incorrect because Azure OpenAI can answer questions but requires prompt engineering and may not use the PDFs directly.

Option D is incorrect because pre-built question answering is for common scenarios, not custom PDFs.

Full explanation →
62
Drag & Dropmedium

Drag and drop the steps to set up Azure AI Content Safety for content moderation into the correct order.

Drag steps to the numbered slots on the right, or tap a step then tap a slot.

Steps
Order

Why this order

First create the resource, get credentials, call the API, analyze, then act.

Full explanation →
63
MCQmedium

A company uses Microsoft Copilot for Microsoft 365 to automate email responses. They want to ensure that the Copilot responses comply with their data governance policies and do not expose sensitive information. What should they configure?

A.Microsoft Entra ID conditional access policies
B.Microsoft Purview Data Loss Prevention (DLP) policies
C.Microsoft Intune app protection policies
D.Microsoft Sentinel analytics rules
AnswerB

Purview DLP can prevent sensitive data from being shared in Copilot responses.

Why this answer

Microsoft Purview Data Loss Prevention (DLP) policies are designed to detect and prevent the accidental sharing of sensitive information, such as credit card numbers or personally identifiable information (PII), across Microsoft 365 services. By configuring DLP policies, the company can scan Copilot-generated email responses for sensitive data patterns and block or warn before the email is sent, ensuring compliance with data governance policies.

Exam trap

The trap here is that candidates often confuse data governance and content inspection with access control (Entra ID) or endpoint security (Intune), leading them to select a policy that governs who can access data rather than what data is allowed to be shared.

How to eliminate wrong answers

Option A is wrong because Microsoft Entra ID conditional access policies control authentication and access to applications based on user, device, or location conditions, but they do not inspect or govern the content of Copilot-generated email responses for sensitive data. Option C is wrong because Microsoft Intune app protection policies manage how data is handled within mobile apps (e.g., copy/paste restrictions, encryption at rest), but they do not scan or enforce content-level rules on email responses generated by Copilot. Option D is wrong because Microsoft Sentinel analytics rules are used for security information and event management (SIEM) to detect threats and anomalies in logs, not to enforce data governance or prevent sensitive data exposure in email content.

Full explanation →
64
MCQhard

You are designing an Azure Cognitive Search solution for a legal document repository. The solution must support search across multiple languages, including English and Spanish. You need to ensure that searches return documents regardless of the language of the query. Which approach should you use?

A.Create a custom analyzer that splits text by language.
B.Use a single language analyzer set to English, and assume Spanish documents are translated.
C.Use the Microsoft analyzer for English and Spanish in a multi-language field.
D.Create separate fields for each language and set the analyzer per field.
AnswerC

Azure Cognitive Search supports multi-language fields with language-specific analyzers.

Why this answer

Option D is correct because using a multi-language analyzer that includes both English and Spanish microsoft analyzers allows the index to handle multiple languages in the same field. Option A is wrong because a single analyzer cannot handle multiple languages. Option B is wrong because a custom analyzer would not be prebuilt for multiple languages.

Option C is wrong because language-specific fields would require the user to know the language of the query.

Full explanation →
65
Multi-Selectmedium

Which TWO actions are recommended to secure an Azure AI Language resource?

Select 2 answers
A.Use managed identities for authentication from Azure services
B.Enable local authentication with access keys
C.Disable public network access and use a private endpoint
D.Allow all Azure services through the firewall
E.Enable anonymous access for public APIs
AnswersA, C

Managed identities eliminate the need for hard-coded credentials.

Why this answer

Managed identities allow Azure AI Language resources to authenticate to other Azure services (like Azure Storage or Azure Key Vault) without storing credentials in code or configuration. This eliminates the risk of access key leakage and aligns with the principle of least privilege, as the identity is tied to the resource lifecycle and can be granted granular RBAC permissions.

Exam trap

The trap here is that candidates often assume disabling public network access alone is sufficient, but they forget that managed identities are required to replace access keys for authentication from Azure services, making both A and C necessary together.

Full explanation →
66
MCQhard

Refer to the exhibit. You are using Azure OpenAI Service to generate summaries of meeting notes. The current configuration produces summaries that are too short and sometimes omit key points. How should you modify the parameters to get more complete summaries?

A.Increase max_tokens to 300.
B.Decrease temperature to 0.3.
C.Increase frequency_penalty to 0.5.
D.Set top_p to 0.9.
AnswerA

Higher max_tokens allows longer completions.

Why this answer

Option B is correct because increasing max_tokens allows the model to generate longer summaries. Option A is wrong because lowering temperature increases determinism but not length. Option C is wrong because increasing frequency_penalty reduces repetition, not length.

Option D is wrong because changing top_p affects diversity, not length.

Full explanation →
67
MCQhard

Your company uses Azure AI Search for an internal knowledge base. Users complain that searches for 'annual report 2023' return irrelevant results. You analyze the search index and find that the content field contains large blocks of text from PDFs. You need to improve relevance without re-indexing all documents. Which approach should you take?

A.Enable spell correction in the search query
B.Add a custom scoring profile based on term frequency
C.Change the index analyzer to a different language
D.Enable semantic ranking on the search index
AnswerD

Semantic ranking re-ranks results using deep learning models to better match query intent.

Why this answer

Option A is correct because applying Azure AI Search's built-in semantic ranking re-ranks results using language understanding, improving relevance for natural language queries without re-indexing. Option B is wrong because enabling spell correction in the query only fixes typos, not relevance. Option C is wrong because adding a custom scoring profile requires changes to the index definition, which may require re-indexing.

Option D is wrong because changing the analyzer to a different language does not address the core relevance issue.

Full explanation →
68
MCQmedium

You are designing a knowledge mining solution that ingests documents from SharePoint Online and makes them searchable using Azure AI Search. The solution must extract text from images and perform optical character recognition (OCR) on embedded images within PDFs. Which built-in skill should you include in the skillset?

A.OCR skill
B.Translation skill
C.Key phrase extraction skill
D.Entity recognition skill
AnswerA

Extracts text from images and embedded images in PDFs.

Why this answer

Option B is correct because the OCR skill extracts text from images. Option A is for text translation. Option C is for key phrases.

Option D is for entity recognition.

Full explanation →
69
MCQmedium

Your retail company uses Azure AI Face API to identify VIP customers when they enter a store. Due to privacy regulations, you must obtain explicit consent and store minimal biometric data. Which approach meets these requirements?

A.Train a Custom Vision model with images of VIP customers.
B.Use the Face API's Face ID returned upon detection and store it in a database.
C.Store the customer's face image in Azure Blob Storage and compare using the Face API.
D.Create a PersonGroup with each VIP customer, and use the Identify operation.
AnswerD

PersonGroup stores only a template, not the image, aligning with privacy.

Why this answer

Option C is correct because Face API's PersonGroup stores only a numeric ID per face, not the actual image, which minimizes biometric data. Storing images (A) or using Face ID (B) retains more data. Custom Vision (D) is not designed for face identification.

Full explanation →
70
MCQhard

You are responsible for managing costs for multiple Azure AI services in your organization. You notice that provisioned throughput units (PTUs) for Azure OpenAI are not fully utilized. What is the most cost-effective action to optimize spending?

A.Configure auto-scaling to reduce PTUs during low usage
B.Move the resource to a different region with lower pricing
C.Stop the Azure OpenAI service when not in use
D.Reduce the number of provisioned PTUs or switch to pay-as-you-go
AnswerD

Adjusting PTU commitment to actual usage or using token-based billing reduces costs.

Why this answer

Option D is correct because provisioned throughput units (PTUs) represent a fixed capacity commitment. If PTUs are underutilized, the most cost-effective action is to reduce the number of PTUs or switch to the pay-as-you-go (PAYG) model, which charges only for tokens consumed. This directly aligns with cost optimization by eliminating the fixed cost of unused capacity.

Exam trap

The trap here is that candidates confuse PTUs with standard Azure auto-scaling concepts (like VM scale sets) and assume auto-scaling is available for PTUs, when in fact PTUs are a fixed-capacity model that requires manual adjustment or a switch to PAYG for cost optimization.

How to eliminate wrong answers

Option A is wrong because Azure OpenAI does not support auto-scaling for PTUs; PTUs are a fixed, pre-provisioned capacity model, and scaling must be done manually via quota adjustments. Option B is wrong because moving to a different region does not address underutilization of PTUs; regional pricing differences are typically minimal and do not solve the core issue of paying for unused capacity. Option C is wrong because stopping the Azure OpenAI service (e.g., via the Azure portal) is not a supported operation for the resource itself; you can only pause or delete the resource, which would lose all configurations and data, making it impractical for intermittent use.

Full explanation →
71
MCQmedium

You are developing an agentic solution that requires the agent to maintain context across multiple turns in a conversation. Which feature should you use to store and retrieve conversation history?

A.Sessions
B.Threads
C.Vector store
D.Memory
AnswerB

Threads automatically store messages and context across turns.

Why this answer

Threads in Azure AI Agent Service are designed to store conversation history and maintain context. Memory is a generic term, not specific to the service. Vector stores are for embeddings.

Sessions are for containerized apps.

Full explanation →
72
MCQmedium

You are configuring semantic search in Azure AI Search. Based on the exhibit, which field is used as the title field for semantic ranking?

A.my-semantic-config
B.description
C.title
D.prioritizedKeywordsFields
AnswerC

The titleField property specifies the title field.

Why this answer

Option C is correct because in Azure AI Search semantic ranking, the 'title' field is designated as the primary field for semantic ranking. This field provides a concise, high-level summary of the document content, which the semantic ranker uses to understand the document's main topic and improve relevance scoring. The semantic configuration references this field via the 'titleField' parameter in the semantic configuration object.

Exam trap

The trap here is that candidates often confuse the semantic configuration name (e.g., 'my-semantic-config') with the actual field names used in the configuration, leading them to select the configuration name instead of the correct field like 'title'.

How to eliminate wrong answers

Option A is wrong because 'my-semantic-config' is the name of the semantic configuration object, not a field used for semantic ranking; it defines the configuration but is not a data field. Option B is wrong because 'description' is typically used as the content field (via 'contentFields') for semantic ranking, not the title field; the title field must be explicitly set to 'title' or a similarly named field. Option D is wrong because 'prioritizedKeywordsFields' is a separate parameter in the semantic configuration that specifies fields for keyword boosting, not the title field for semantic ranking.

Full explanation →
73
MCQhard

You are developing an agentic solution that uses Azure AI Agent Service with a custom function calling tool. The agent needs to call a function that requires authentication to an external API. How should you securely pass the API key to the function?

A.Hardcode the API key in the function code
B.Use Azure Key Vault to store the API key and reference it in the function
C.Store the API key in an environment variable
D.Pass the API key as part of the agent's system prompt
AnswerB

Key Vault provides secure secret storage and retrieval.

Why this answer

Using Azure Key Vault references in the function code is the secure way to retrieve secrets at runtime without exposing them. Hardcoding is insecure. Environment variables may be exposed.

Passing in the prompt leaks credentials.

Full explanation →
74
MCQeasy

You are building a chatbot using Azure OpenAI Service. The chatbot must not disclose sensitive information such as passwords or credit card numbers. Which Azure AI service should you integrate to filter such content?

A.Azure AI Language
B.Azure AI Bot Service
C.Azure AI Search
D.Azure AI Content Safety
AnswerD

It detects and filters unsafe or sensitive content.

Why this answer

Azure AI Content Safety is the correct service because it provides pre-built content filters that can detect and block sensitive information such as passwords, credit card numbers, and other personally identifiable information (PII) in text and images. It integrates directly with Azure OpenAI Service to apply these filters to both input prompts and output completions, ensuring that sensitive data is not disclosed in chatbot responses.

Exam trap

The trap here is that candidates often confuse Azure AI Language's PII detection feature (which identifies but does not block content) with Azure AI Content Safety's filtering capability, leading them to incorrectly choose Azure AI Language for proactive content blocking.

How to eliminate wrong answers

Option A is wrong because Azure AI Language is a natural language processing service for tasks like sentiment analysis, key phrase extraction, and language understanding, but it does not include built-in content filtering for sensitive data like passwords or credit card numbers. Option B is wrong because Azure AI Bot Service is a platform for building, deploying, and managing chatbots, but it does not provide native content filtering capabilities; it relies on other services like Content Safety for such functionality. Option C is wrong because Azure AI Search is a search-as-a-service solution for indexing and querying data, not a content safety or filtering service, and it cannot detect or block sensitive information in real-time chatbot interactions.

Full explanation →
75
MCQeasy

Your company uses Azure AI Language to analyze customer feedback. You need to extract key phrases from reviews in multiple languages. Which feature should you use?

A.Named entity recognition
B.Language detection
C.Key phrase extraction
D.Sentiment analysis
AnswerC

Key phrase extraction extracts the main concepts from text.

Why this answer

Key phrase extraction is the correct feature for identifying main points in text. Sentiment analysis detects positive/negative tone, entity recognition identifies named entities, and language detection identifies the language.

Full explanation →

Page 1 of 14

Page 2

All pages

1234567891011121314

Practice AI-102 by domain

Target a specific domain to shore up weak areas.

Implement an agentic solutionImplement computer vision solutionsImplement knowledge mining and information extraction solutionsImplement image and video processing solutionsImplement natural language processing solutionsImplement generative AI solutionsImplement agentic AI solutionsImplement knowledge mining and document intelligence solutionsPlan and manage an Azure AI solutionImplement content moderation solutions
See all domains with question counts →