Microsoft Practice Questions, Discussions & Exam Topics by our Authors

Which metric can you use to evaluate a classification model?

When evaluating a classification model, the correct metric needs to focus on assessing how well the model differentiates between the classes, as well as its predictive accuracy. Let's break down each of the options: A) True Positive Rate (TPR) - Explanation: TPR, also known as recall or sensitivity, measures the proportion of actual positives that are correctly identified by the model. In classification problems, it is crucial to evaluate how well the model is identifying the relevant instances of the target class. - Why selected: This is a key metric for classification models, especially in imbalanced classes, as it helps assess the performance in detecting the positive class. - Scenario: Use when the goal is to minimize false negatives (e.g., in medical diagnosis where missing a positive case could have significant consequences). B) Mean Absolute Error (MAE) - Explanation: MAE is a metric used for regression tasks, measuring the average magnitude of the errors in a set of predictions, without considering their direction. - Why rejected: MAE is not suitable for classification tasks, as it is designed for continuous variables, not categorical outcomes. It doesn't provide meaningful information about classification accuracy. - Scenario: MAE should be used for regression models, not classification. C) Coefficient of Determination (R²) - Exp...

Author: StarryEagle42 · Last updated Jul 13, 2026

You have an Azure subscription that contains the resources shown in the following table. You need to configure storage1 to regenerate key...

To configure storage1 to regenerate keys automatically every 90 days in an Azure subscription, we need to focus on how to manage the keys and their rotation. Let's evaluate the cmdlet options available: Key Factors: - The goal is to automatically regenerate keys for storage1 every 90 days. - Storage account keys management is typically handled by Azure Storage Management Policies. These policies can automate tasks like key regeneration. Analysis of Options: A) Add-AzKeyVaultflanagedStorageAccount - This cmdlet is not a valid cmdlet or part of the Azure PowerShell module. It seems like a typo or an invalid cmdlet. - Rejected because it is incorrect and doesn't exist in the Azure PowerShell cmdlets. B) Set-AzStorageAccountManagementPolicy - This cmdlet is used to configure and manage storage account management policies. - Management policies allow the automation of various tasks, such as object lifecycle management and key regeneration. It could be used for configuring how storage accounts handle key rotation automatically. - Selected option because this is the ...

Author: Daniel · Last updated Aug 15, 2026

Which two components can you drag onto a canvas in Azure Machine Learning designer? Each correct answer presents a complete solu...

In Azure Machine Learning Designer, the correct components to drag onto a canvas are those that can be used to construct and design a machine learning workflow. Let's analyze the given options: A) Dataset - Explanation: A dataset is a collection of data used for training and testing machine learning models. In Azure Machine Learning Designer, datasets are often the starting point for building workflows, as they provide the input data for model training or evaluation. - Why selected: You can drag datasets onto the canvas to feed them into modules for model building or evaluation. It's essential to specify your input data for model training or testing. - Scenario: Use when you want to work with data that will be used for preprocessing, training, or evaluation in your machine learning workflow. B) Compute - Explanation: Compute in Azure Machine Learning refers to the infrastructure (such as virtual machines or compute clusters) used to run the machine learning models or pipelines. While you need compute resources for training models or running experiments, these resources are not directly dragged onto the canvas. - Why rejected: Compute resources are managed at the workspace level and assigned to a pipeline or experiment, but you do not drag them directly onto the canvas in the designer interface. - Scenario: Compute is used for running experiments or training models, but it is not an object you drag onto the canvas in Azure Machine Learning Designer. C) Pipeline - Explanation: A pipeline is a collection of modules and steps that represent the...

Author: Emma Brown · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure subscription that contains the key vaults shown in the following table. The subscription contains the users shown in the following table. On June 1, you perform the following actions: * Delete a key named key1 from KeyVault1. * Delete a secret named secret1 from KeyVault2. For e...

Author: Aarav · Last updated Aug 15, 2026

You need to create a training dataset and validation dataset from an existing dataset. Which module in the ...

To create a training dataset and a validation dataset from an existing dataset in Azure Machine Learning Designer, the correct module should allow you to split the dataset into multiple subsets. Let’s review the options: A) Select Columns in Dataset - Explanation: The "Select Columns in Dataset" module is used to choose specific columns from a dataset based on user-defined criteria. This is helpful when you need to filter out unnecessary features or select only relevant columns for your analysis or model training. - Why rejected: This module doesn't split the dataset into different subsets like training and validation sets. It focuses on selecting or removing columns from a dataset, which is not the intended function for creating training and validation datasets. - Scenario: Use when you need to select specific features (columns) from your dataset, not for splitting the dataset. B) Add Rows - Explanation: The "Add Rows" module is used to append additional rows (data) to an existing dataset. This could be used to augment or expand a dataset by adding more data, but it doesn’t help with creating training and validation subsets from a single dataset. - Why rejected: This module does not split data into training or validation subsets. It is used for adding data rather than splitting it into distinct sets. - Scenario: Use when you want to append more rows to a dataset, not when creating training and validation datasets. ...

Author: IronLion88 · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure Active Directory (Azure AD) tenant that contains two administrative units named AU1 and AU2. Users are assigned to the administrative units as shown in the following table. Users are assigned the roles shown in the following table. For each of the fo...

Author: Liam · Last updated Aug 15, 2026

DRAG DROP - Match the types of machine learning to the appropriate scenarios. To answer, drag the appropriate machine learning type from the column on the left to its scenario on the right. Each machine learning type may be used once...

Absolutely! Let’s break it down carefully so the reasoning is crystal clear: --- 1. Predict how many minutes late a flight will arrive Problem type: You are trying to predict a numeric value (minutes late). ML type: Regression Regression is used when the output is continuous, like prices, temperatures, or counts. Example: If snowfall = 5 inches → predict 20 minutes late. --- 2. Segment customers into different groups to support marketing Problem type: You are trying to group customers based on their behavior or attributes. ML type: Clustering Clustering is unsupervised learning, meaning there’s no labeled output. Example: Group customers into “High-value,” “Occasional,” and “Potential” clusters for marketing campaigns. --- 3. Predict whether a student will complete ...

Author: Emma · Last updated Jul 13, 2026

You have an Azure subscription that contains an Azure key vault named Vault1 and a virtual machine named VM1. VM1 has the Key Vault VM extension installed. For Vault1, you rotate th...

To determine what will be updated automatically on VM1 after rotating keys, secrets, and certificates in Vault1, let's analyze the situation and how the Key Vault VM extension works. Key Factors: - Azure Key Vault VM Extension: This extension is designed to allow virtual machines to securely retrieve secrets, certificates, and keys from an Azure Key Vault. The extension can also automatically update the secrets and certificates on the virtual machine when changes occur in the Key Vault. - Key Vault Rotation: When you rotate keys, secrets, or certificates in Key Vault, the VM extension will automatically update only the relevant resources that it is configured to track and refresh. - Keys, Secrets, and Certificates: Key Vault stores and manages keys, secrets, and certificates, but the automatic updates on the VM depend on what is being tracked by the VM extension. Analysis of Options: A) The keys only - The Key Vault VM extension is capable of tracking and automatically updating only the secrets and certificates that are used by the VM. However, it does not automatically update keys. - Rejected because the extension does not automatically update the keys on the VM. B) The secrets only - The Key Vault VM extension can automatically update secrets on the VM when they are rotated in the Key Vault. - Partially correct, but it misses t...

Author: Maya · Last updated Aug 15, 2026

SNAPSHOT - To complete the sentence, select the appropriate option in the answer area. ___________________is the calculated p...

Correct answer: B) Confidence ✅ Explanation (AI-900 context): Confidence is the calculated probability that a model’s prediction (for example, an image classification) is correct. Example: “Cat – 0.92 confidence” means the model is 92% sure the image is a cat. Accuracy measures overa...

Author: Suresh · Last updated Jul 13, 2026

SNAPSHOT - To complete the sentence, select the appropriate option in the answer area. Hot Area:

The correct answer is: C) reliability and safety ✅ Explanation: Reliability and safety ensures that an AI system behaves predictably and safely, including avoiding predictions when critical input data is missing, incomplete, or unusual. ...

Author: Oscar · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure SQL database named DB1 that contains a table named Tablet. You need to configure DB1 to meet the following requirements: * Sensitive data in Table1 must be identified automatically. * Only the first character and last character of the sensitive data must be displayed in query results. W...

Author: Max · Last updated Aug 15, 2026

SNAPSHOT - To complete the sentence, select the appropriate option in the answer area. Hot Area: Ensuring that the numeric variables in training...

Correct answer: B) Feature engineering ✅ Explanation: Ensuring numeric variables are on a similar scale (for example, normalization or standardization) is a feature engineering task. Data ingestion is about collect...

Author: Lina Zhang · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure subscription that contains two users named User1 and User2 and the blob containers shown in the following table. Policy1 is configured as shown in the following exhibit. You assign the roles for storage1 as shown in the following table. The storage1 account has the following shared access signature (SAS) named SAS1: * Allowed services: Blob * Allowed resource types: Container * Allowed permissions: Read, Write, List, Add, Create * Blob versioning permissions: enables deletion of versions * Allowed blob index...

Author: Joseph · Last updated Aug 15, 2026

SNAPSHOT - To complete the sentence, select the appropriate option in the answer area. Assigning classes to images before training a clas...

The correct answer is: D) labeling ✅ Here’s why: Labeling is the process of assigning the correct output (class, category, or value) to each data point in your dataset. In image classification, this means tagging each image with the class it belongs to before training the model. Evaluation happens after the ...

Author: Zara · Last updated Jul 13, 2026

SNAPSHOT - Your on-premises network contains the servers shown in the following table. You have an Azure subscription that contains multiple virtual machines that run either Windows Server 2019 or SLES. You plan to implement adaptive application controls in Microsoft Defender for Cloud. Which operating systems an...

Author: Aarav2020 · Last updated Aug 15, 2026

SNAPSHOT - You have an Azure Machine Learning model that predicts product quality. The model has a training dataset that contains 50,000 records. A sample of the data is shown in the following table. For each of the following statements, select Yes...

The correct answer is: C) Yes, Yes, No ✅ Explanation: 1. Mass (kg) is a feature → Yes Features are the input variables used to predict a target. Mass is an input measurement. 2. Quality Test is a label → Yes L...

Author: Daniel · Last updated Jul 13, 2026

You have an Azure subscription that contains the resources shown in the following table. Both VM1 and VM2 connect to VNET1 and are configured to use NSG1. You n...

To ensure that only VM1 and VM2 can access DB1, we need to implement a strategy that restricts access based on network and IP configurations while adhering to the principle of least privilege. Let's evaluate each option: Key Factors to Consider: - VM1 and VM2 are connected to VNET1 and are governed by NSG1 (Network Security Group). - DB1 should be accessible only from VM1 and VM2, so we need to restrict access at the network level. - We need to block access from other resources while allowing specific VMs within the same Virtual Network to connect to DB1. Analysis of Options: A) For NSG1, configure a rule that has a service tag. - Service tags are used to simplify network security rules by grouping IP ranges for specific Azure services, such as Azure SQL Database, Azure Storage, etc. However, service tags are designed to control access to Azure services, not specifically between individual virtual machines within a Virtual Network (VNET). - While service tags are helpful for managing access to Azure services, they are not ideal for restricting access to a specific VM or set of VMs. - Rejected because it won't control access specifically between VM1 and VM2 and DB1. B) Add the IP address range of VNET1 to the Firewall settings of DB1. - Adding the IP address range of VNET1 to the firewall settings of DB1 would allow all resources in VNET1, including VM1, VM2, and any other resources within VNET1, to access DB1. - This is not restrictive enough b...

Author: Rahul · Last updated Aug 15, 2026

SNAPSHOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Ea...

The correct answer is: B) No, No, Yes Explanation: 1. You train a regression model by using unlabeled data? ❌ No — Regression requires labeled data (known input–output pairs). 2. The classification technique is used to predict sequential numerical data over time? ❌ No — Predicting sequential...

Author: Andrew · Last updated Jul 13, 2026

You have an Azure AD tenant that contains a user named User1. You purchase an app named App1. User1 needs to publish App1 by using Azure A...

To publish an application through Azure AD Application Proxy, the user needs the appropriate role that grants the permissions to configure and publish the app. Key Factors to Consider: - Azure AD Application Proxy is a service that allows organizations to securely publish internal apps to external users. - A user needs the correct role to publish and manage apps through Azure AD Application Proxy. - The role must provide permissions specifically for managing applications, rather than broader roles like identity management or security. Analysis of Options: A) Cloud application administrator - This role has permissions to manage and configure applications in Azure AD, which includes configuring Application Proxy for apps. - It allows the user to add, configure, and publish apps via the Azure AD Application Proxy service, which is exactly what User1 needs to do to publish App1. - Selected option because it provides the necessary permissions for publishing and managing apps using Application Proxy. B) Application administrator - The Application administrator role allows users to manage application registrations and enterprise applications in Azure AD. This includes tasks such as managing user access and configuring application properties. - However, it doesn't explicitly give permissions for publishing applications through Azure AD Application Proxy. The role is more focused on app registrations and a...

Author: Aditya · Last updated Aug 15, 2026

Which two actions are performed during the data ingestion and data preparation stage of an Azure Machine Learning process? Each correct answer presents part...

Explanation of each option: - A) Calculate the accuracy of the model: This action typically occurs after a model has been trained and evaluated, not during the data ingestion and preparation stage. At the data ingestion and preparation stage, you're working with raw data and not yet evaluating the model's performance. Accuracy calculation is part of the model evaluation phase, which follows the training process. - B) Score test data by using the model: Scoring data refers to applying the trained model to test data to make predictions. This happens after data preparation, model training, and evaluation. Therefore, it is not a part of data ingestion and preparation. - C) Combine multiple datasets: This action is a key part of the data ingestion and preparation stage. When working with multiple datasets, they are often combined (or joined) in order to ensure that all relevant features are included for training the model. This could involve merging different data sources to create a unifi...

Author: Sofia · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure AD tenant named contoso.com that contains the users shown in the following table. You add enterprise applications to contoso.com as shown in the following table. You need to identify which users can grant admin consent for App1 and App2. Which users should you identif...

Author: VioletCheetah55 · Last updated Aug 15, 2026

You need to predict the animal population of an area. Which Azure Machine Learning type should you ...

Explanation of each option: - A) Regression: Regression is used for predicting continuous numeric values, such as predicting the population of an animal in a certain area. Since predicting the animal population involves forecasting a numerical value (the population count), regression is the most appropriate choice. It helps model the relationship between the input features (such as environmental factors, animal behavior, etc.) and the continuous target variable (population). - B) Clustering: Clustering is a technique used for grouping similar data points together based on their characteristics. It is unsupervised learning, where no explicit labels or target values are involved. While clustering could be useful in identifying patterns or groups of animals in the area, it is not appropriate for predicting a specific numeric value like populatio...

Author: Harper · Last updated Jul 13, 2026

DRAG DROP - Your network contains an on-premises Active Directory domain named contoso.com. The domain contains a user named User1. You have an Azure subscription that is linked to an Azure Active Directory (Azure AD) tenant named contoso.com. The tenant contains an Azure Storage account named storage1. Storage1 contains an Azure file share named share1. Currently, the domain and the tenant are not integrated. You need to ensure that User1 can access share1 by u...

Author: Deepak · Last updated Aug 15, 2026

Which two languages can you use to write custom code for Azure Machine Learning designer? Each correct answer presents a complete so...

Explanation of each option: - A) Python: Python is one of the most popular languages for data science and machine learning. Azure Machine Learning designer supports Python for writing custom code, including custom transformations, model training, and evaluation. It is a widely used language for data analysis, manipulation, and creating machine learning models, making it an ideal choice for custom code. - B) R: R is another programming language supported by Azure Machine Learning designer, particularly for statistical analysis and data visualization. It is widely used in academic and research settings, especially for tasks related to data analysis, modeling, and statistical computation. R is supported for writing custom code in the Azure ML designer as well. - C) C: C is not a language typically used within Azure Machine Learning designer for writing custom code. Azure ML is designed primarily around Python and ...

Author: Manish · Last updated Jul 13, 2026

SNAPSHOT - You have a Microsoft Sentinel deployment. You need to connect a third-party security solution to the deployment. The third-party solution will send Common Event Format (CEF)-formatted messages. What should you include in the solution? To...

Author: Ethan Smith · Last updated Aug 15, 2026

SNAPSHOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Ea...

The correct answer is: D) Yes, No, No Explanation: 1. For a regression model, labels must be numeric? ✅ Yes — Regression predicts continuous numeric values. 2. For a clustering model, labels must be used? ❌ No — Clustering is a...

Author: Michael · Last updated Jul 13, 2026

You have an Azure subscription that contains the virtual machines shown in the following table. Which c...

To determine which computers will support File Integrity Monitoring (FIM), we need to understand the requirements and capabilities of the environment. Key Factors: - File Integrity Monitoring (FIM) is a security feature that monitors changes to files or directories, typically used to detect unauthorized modifications to important files. - In Azure, FIM is commonly associated with Azure Security Center, Microsoft Defender for Identity, or Azure Sentinel for enhanced security monitoring. - For FIM to be supported, the operating system, Azure configuration, and specific security settings must be compatible with FIM requirements. Analysis of Options: Without seeing the actual table of computers (since it’s not provided), we can still assess the general criteria for FIM support based on typical conditions: A) Computer2 only - If Computer2 is the only machine configured correctly for FIM, such as using the appropriate Windows Server version or Linux distribution with FIM capabilities enabled, this option could be correct. - However, we can't assume without further details that only Computer2 supports FIM. B) Computer1 and Computer2 only - This option would imply that Computer1 and Computer2 are the only machines that meet the requirements for FIM, such as having a compatible OS or security ...

Author: Ishaan · Last updated Aug 15, 2026

Your company wants to build a recycling machine for bottles. The recycling machine must automatically identify bottles of the correct shape and reject a...

Explanation of each option: - A) Anomaly detection: Anomaly detection is used to identify unusual patterns or outliers in data. While it can be useful for identifying rare or unexpected behaviors, it is not ideal for identifying objects with specific shapes. The problem described involves identifying bottles based on their shape, which requires understanding the visual characteristics of objects, not detecting anomalies in data. - B) Conversational AI: Conversational AI involves building systems like chatbots or virtual assistants that understand and generate human language, typically through text or voice. This is not applicable to the problem at hand, which requires recognizing physical objects (bottles) based on their shape, not engaging in a conversation or processing language. - C) Computer vision: Computer vision is the most appropriate AI workload for this scenario. It is designed to enable machines to interpret and process visual information from the world, such as ...

Author: Aarav2020 · Last updated Jul 13, 2026

SIMULATION - The developers at your company plan to create a web app named App28681041 and to publish the app to https://www.contoso.com. You need to perform the following tasks: * Ensure that App28681041 is registered to Azu...

To complete the tasks provided for App28681041, let's break down the requirements: 1. Ensure that App28681041 is registered to Azure AD. - App28681041 needs to be registered in Azure Active Directory (Azure AD) to authenticate and integrate with Azure AD for identity management and permissions. 2. Generate a password for App28681041. - After the app is registered, you need to generate a client secret (password) for App28681041, so it can authenticate securely with Azure AD. Key Steps for the Tasks: 1. Register the application in Azure AD. - To register the app, you need to go to Azure Active Directory > App registrations and either create a new app or check if the app is already registered. For this scenario, you would register App28681041 if it hasn't been done yet. 2. Generate a client secret for the app. - After registering the app, you will need to generate a client secret by navigating to the Certificates & secrets section of the app registration page in the Azure portal. This will be the password you use to authenticate the app. Evaluation of Potential Options (assuming options are about navigating the Azure portal): A) Go to the Azure AD App Registrations section and create a new application - This step is correct for registering an app in Azure AD, which satisfies the first task. - However, this option alone doesn't cover generating a password (client secret) f...

Author: Grace · Last updated Aug 15, 2026

SNAPSHOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Ea...

Let’s carefully go through each statement about Custom Vision for the AI-900 exam: 1. “For AI900, When creating an object detection model in the Custom Vision service, You must choose a classification type of either Multilabel or Multiclass?” False. Object detection models are different from classification models. You do not choose Multiclass or Multilabel for object detection; that choice is only for classification models. ✅ 2. “You can create an object detection model in the Custom Vision service to find the...

Author: Liam123 · Last updated Jul 13, 2026

You have an Azure AD tenant that contains the users shown in the following table. You need to ensure that the users cannot create app passwords. The solution must ensure th...

To address the requirement where users cannot create app passwords, while ensuring that User1 can continue to use the Mail and Calendar app, we need to focus on how Azure AD controls app passwords and multi-factor authentication (MFA). Let's break down each option: Key Factors: - App passwords are used for apps that do not support modern authentication or MFA. When MFA is enabled, users can create app passwords to use with older apps. - We want to prevent the creation of app passwords, but still ensure User1 can use the Mail and Calendar app, which might require app passwords if it's not integrated with modern authentication. Analysis of Options: A) Assign User1 the Authentication Policy Administrator role. - The Authentication Policy Administrator role allows users to manage authentication policies, including the ability to configure multi-factor authentication (MFA) settings and policies for users in Azure AD. - While this role could help manage authentication policies, it doesn't directly control whether users can create app passwords. - Rejected because this role doesn't directly prevent the creation of app passwords or achieve the goal described. B) Enable Azure AD Password Protection. - Azure AD Password Protection helps protect against weak or commonly used passwords. It is designed to improve password security but does not directly control the ability to create app passwords or manage MFA settings. - Rejected because this option does not prevent the creation of app passwords or address the MFA-related issues. C) Configure a multi-factor aut...

Author: RadiantPhoenixX · Last updated Aug 15, 2026

In which two scenarios can you use the Form Recognizer service? Each correct answer presents a complete solution. ...

Explanation of each option: - A) Extract the invoice number from an invoice: Form Recognizer is designed to extract structured data from forms, such as invoices, receipts, and other document types. It can automatically identify and extract specific fields like the invoice number, date, total, and line items. Therefore, extracting the invoice number from an invoice is a scenario that fits well with the capabilities of Form Recognizer. - B) Translate a form from French to English: Form Recognizer does not perform language translation. Its primary function is to analyze the structure of forms and extract data from them. Translation of a form from one language to another (e.g., from French to English) would require a service like Azure Translator, not Form Recognizer. Therefore, this is not a valid scenario for Form Recognizer. - C) Find image of product in a catalog: Form Recognizer is focused on extracting data from structured documents ...

Author: Max · Last updated Jul 13, 2026

SNAPSHOT - Select the answer that correctly completes the sentence. Counting the number of animals in an area based on a v...

Let’s carefully analyze the question. Task: Counting animals in an area based on a video feed. A) Forecasting ❌ Forecasting predicts future values based on historical data. Counting animals in real time is not prediction. B) Computer vision ✅ Computer vision analyzes images or video to detect, count, or tr...

Author: Sophia Clark · Last updated Jul 13, 2026

DRAG DROP - You have an Azure subscription named Sub1 that contains the storage accounts shown in the following table. The storage3 storage account is encrypted by using customer-managed keys. You need to enable Microsoft Defender for Storage to meet the following requirements: * The storage1 and storage2 accounts must be included in the Defender for Storage protections. * The storage3 account must be exclude...

Author: Mia · Last updated Aug 15, 2026

SNAPSHOT - You have a database that contains a list of employees and their photos. You are tagging new photos of the employees. For each of the following statements select Yes if the statement is t...

Absolutely! Let’s go statement by statement with explanations for clarity: --- 1. “The Face service can be used to perform facial recognition for employees?” ✅ Yes Explanation: The Azure Face service can detect, identify, and verify faces. This makes it ideal for employee authentication scenarios. Example: You can enroll employee faces in the system and then verify their identity when they log in or enter a building. --- 2. “The Face service will be more accurate if you provide more sample photos of each employee from different angles?” ✅ Yes Explanation: AI models perform better with diverse training data. Providing multiple images of an employee ...

Author: Abigail · Last updated Jul 13, 2026

You have an Azure subscription that contains the virtual machines shown in the following table. You are configuring Microsoft Defender for Servers. You plan to enable adaptive application controls to create an allowlist of known-safe ...

To determine which virtual machines support the use of adaptive application controls in Microsoft Defender for Servers, we need to understand some key factors related to adaptive application controls, which are a part of Defender for Servers' security offerings. These controls allow you to create an allowlist of known-safe applications for your virtual machines. To use adaptive application controls, your virtual machines must meet certain prerequisites: 1. Operating System Support: Adaptive application controls primarily work on Windows Server 2016 and later and Linux distributions that are supported by Microsoft Defender for Servers. 2. Microsoft Defender for Servers: The virtual machine should have Microsoft Defender for Servers enabled, as this service allows you to configure application control policies. 3. Availability of Application Control Features: Adaptive application control may not work in some configurations where custom security settings or unsupported OS versions are used. Analysis of Options: - VM1: If VM1 is running an unsupported operating system version or lacks Microsoft Defender for Servers, it won't support adaptive application controls. - VM2: If VM2 is running a supported operating system (like Windows Server 2016 or later, or supported Linux distributions), it will be eligible for adaptive application controls. - VM3: ...

Author: Layla · Last updated Aug 15, 2026

You need to develop a mobile app for employees to scan and store their expenses while travelling. W...

Explanation of each option: - A) Semantic segmentation: Semantic segmentation is a computer vision technique used to partition an image into regions that are semantically meaningful, such as separating different objects or areas within an image. It is typically used in scenarios like identifying roads in driving images or segmenting regions in medical images. For the task of scanning and storing expenses, semantic segmentation is not suitable, as the primary need is to extract text from receipts, not segmenting objects or regions in the image. - B) Image classification: Image classification involves categorizing an image into a predefined class or category. This technique is useful when you want to classify images based on general features, like identifying whether an image contains a dog, a cat, or a car. However, for scanning expenses, the task is not to classify the image but rather to extract specific textual data (such as the amount, date, or merchant) from receipts, making image classification inappropriate for this use case. - C) Object detection: Object detection is used to identify and...

Author: Vikram · Last updated Jul 13, 2026

You have an Azure subscription. The subscription contains a virtual network named VNet1 that contains the subnets shown in the following table. The subscription contains the function apps shown in t...

To determine which function app's outbound traffic is controlled by NSG1, we need to consider the following key factors: 1. Network Security Groups (NSGs): - NSGs control network traffic to and from resources in Azure Virtual Networks (VNets). - NSGs can be applied to subnets or network interfaces (NICs). - If an NSG is applied to a subnet, it controls the traffic to and from all resources in that subnet. 2. Function Apps and Outbound Traffic: - Function apps in Azure are typically hosted in an App Service Environment (ASE) or a VNet Integration scenario. If they are in an ASE or integrated with the VNet, their traffic may pass through a subnet that has an NSG attached. - If the function app is in a subnet with NSG1 applied, outbound traffic will be controlled by NSG1. Key Steps to Determine the Correct Answer: 1. Check which subnets the function apps are placed in. - The function apps would inherit the NSG rules if they are deployed within the same subnet as the NSG. - The outbound traffic of a function app will be controlled by an NSG if the app resides in a subnet with that NSG. 2. Which Function Apps are in Subnets with NSG1? - We...

Author: RadiantJaguar56 · Last updated Aug 15, 2026

SNAPSHOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Ea...

Absolutely! Let’s go statement by statement with clear explanations: --- 1. “The Custom Vision service can be used to detect objects in an image?” ✅ Yes Explanation: Custom Vision supports object detection, which allows the model to identify what objects are present in an image and where they are located (via bounding boxes). Example: Detecting cars, people, or weeds in an image. --- 2. “The Custom Vision service requires that you provide your own data to train the model?” ✅ Yes Explanation: Custom Vision is a customizable AI service, meaning it does not c...

Author: Harper · Last updated Jul 13, 2026

You have a Microsoft Entra tenant named contoso.com. You collaborate with a partner organization that has a Microsoft Entra tenant named fabrikam.com. You need to create an allow list of cloud apps from fabrikam.com that can be used ...

To create an allow list of cloud apps from fabrikam.com that can be used by the users in contoso.com, you need to manage cross-tenant access and configure external collaboration settings to control which external apps can be accessed. Let's evaluate the options based on the requirements: 1. Option A: From Inbound access settings in Cross-tet access settings, configure the B2B direct connect settings. - This option is used to manage inbound access from other organizations to your own tenant (contoso.com). However, this option is typically used for allowing external organizations' apps to access resources within your tenant, which is the opposite of the requirement. You are looking to control which apps from an external organization (fabrikam.com) can be accessed by your users, not the other way around. 2. Option B: From External collaboration settings, configure the Collaboration restrictions settings. - This option is used to control collaboration settings such as whether users from other organizations (like fabrikam.com) can be invited as guests to your tenant (contoso.com). However, this option doesn't specifically manage which apps from the external tenant can be used by your users. It's more about controlling the overall access of exter...

Author: Vikram · Last updated Aug 15, 2026

You are processing photos of runners in a race. You need to read the numbers on the runners' shirts to identity the runners in th...

Explanation of each option: - A) Facial recognition: Facial recognition is used to identify or verify individuals based on facial features. It is suitable when the goal is to recognize a person's face and match it to a known identity. However, in this scenario, the goal is to read numbers on the runners' shirts, not to recognize their faces, making facial recognition irrelevant for this task. - B) Optical character recognition (OCR): OCR is specifically designed to extract text from images, making it the ideal solution for this task. Since the goal is to read the numbers on the runners' shirts, OCR can recognize the digits and convert them into machine-readable text. This technique is widely used for tasks like scanning printed documents, reading license plates, and extracting text from images, which directly applies to the scenario of reading shirt numbers. - C) Image classification: Image classification involves categorizing an image into predefined ...

Author: Noah · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure subscription that contains the virtual networks shown in the following table. NSG1 rules restrict access to the internet from Subnet3. The subscription contains the function apps shown in the following table. Virtual network integration has the default settings. You need to configure network access for App1 and App2 to meet the following requirements: * Deny inbound access to App1 from Subnet1 and allow inbound access from Subnet...

Author: Kai · Last updated Aug 15, 2026

DRAG DROP - Match the types of machine learning to the appropriate scenarios. To answer, drag the appropriate machine learning type from the column on the left to its scenario on the right. Each machine learning type may be used once...

The correct answer is: A) Image classification, Object detection, Semantic segmentation ✅ Here’s why: 1. Separate images of polar bears and brown bears → Image classification You’re labeling the entire image as either a polar bear or a brown bear. 2. Determine the location of a bear in a photo → Object detection ...

Author: Maya2022 · Last updated Jul 13, 2026

You need to meet the identity and access requirements for Group1. What should you do?

To determine the correct solution for meeting the identity and access requirements for Group1, let's evaluate each option in detail. The goal is to manage group memberships and control access effectively based on Group1's membership and requirements. Key Concepts: - Membership rules: Membership rules in Azure Active Directory (Azure AD) groups, particularly dynamic membership rules, are used to automatically assign members to a group based on specific attributes (e.g., user or device properties). - Group types: The type of group determines the capabilities. Security groups and Microsoft 365 groups are the most common types, with different functionalities such as collaboration tools in Microsoft 365 groups. - Assigned membership: With assigned membership, members are manually added to a group. Evaluating the Options: 1. Option A: Add a membership rule to Group1. - This option is useful if Group1 is intended to have dynamic membership based on specific attributes, such as user properties or device attributes. If the requirement is to automate membership based on certain conditions (like location, department, or other attributes), adding a membership rule to Group1 would be a valid solution. This approach works well if the group is dynamic. 2. Option B: Delete Group1. Create a new group named Group1 that has a group type of Microsoft 365. Add users and devices to the group. - This option would be relevant if Group1 needs to be a Microsoft 365 group (which provides collaboration features such as shared mailboxes, calendars, and other Office 365 services). However, it does not address whether the requirement is related to membership automation or if the group type needs to be changed. Deleting and recreating Group1 should only be done if there is a need for a new group type or additional services not supported by the current type. 3. Option C: Modify the memb...

Author: Leah · Last updated Aug 15, 2026

You use drones to identify where weeds grow between rows of crops to send an instruction for the removal of the weeds. ...

Explanation of each option: - A) Object detection: Object detection is used to identify and locate specific objects within an image. In the context of this scenario, drones would be used to identify the weeds between crop rows, which involves detecting and locating the weeds as distinct objects. Object detection is often used for tasks like recognizing animals, vehicles, or other specific objects, making it a suitable choice for identifying weeds in a field. The drones would identify where the weeds are and trigger an action (like sending an instruction to remove them), making object detection a strong candidate for this scenario. - B) Optical character recognition (OCR): OCR is used to extract text from images, such as scanning printed documents or reading numbers from a license plate. It is not designed for identifying or locating objects like weeds in images, so OCR is not appropriate in this case. - C) Scene segmentation: Scene segmentation is used to divide an image into different region...

Author: Isabella · Last updated Jul 13, 2026

SNAPSHOT - You need to ensure that the Azure AD application registration and consent configurations meet the identity and access requirements. What should you use in the Azure portal? To answer, select the appropr...

Author: Layla · Last updated Aug 15, 2026

DRAG DROP - Match the facial recognition tasks to the appropriate questions. To answer, drag the appropriate task from the column on the left to its question on the right. Each task may be used once, more tha...

Let’s carefully check each one step by step. We have the questions in order: 1. Do two images of a face belong to the same person? → That’s Verification 2. Does this person look like other people? → That’s Similarity 3. Do all the faces belong together? → That’s Grouping 4. Who is this person in this group of people? → That’s Iden...

Author: SilverBear · Last updated Jul 13, 2026

SNAPSHOT - You have an Azure subscription that contains the resources shown in the following table. VNet1 contains the subnets shown in the following table. You plan to use the Azure portal to deploy an Azure firewall named AzFW1 to VNet1. Which resource group and subnet can you use...

Author: NightmareDragon2025 · Last updated Aug 15, 2026

DRAG DROP - Match the types of computer vision workloads to the appropriate scenarios. To answer, drag the appropriate workload type from the column on the left to its scenario on the right. Each workload type may be used once, ...

The correct answer is: A) Facial recognition, Optical character recognition (OCR), Object detection Why: Identify celebrities in images → Facial recognition (detecting and identifying known people) ...

Author: Matthew · Last updated Jul 13, 2026

You have an Azure subscription that contains a storage account named storage1 and a virtual machine named VM1. VM1 is connected to a virtual network named VNet1 that contains one subnet and uses Azure DNS. You need to ensure that VM1 connects to st...

To ensure that VM1 connects to storage1 using a private IP address, we need to configure private connectivity between the virtual machine and the storage account. The solution should minimize administrative effort and ensure that the traffic between VM1 and storage1 is routed privately, rather than using the public internet. Evaluating the Options: 1. Option A: For storage1, disable public network access. - Disabling public network access for storage1 would prevent VM1 from accessing the storage account over the public internet. However, this alone does not provide a private connection between VM1 and storage1. Without a private endpoint or DNS resolution, VM1 may not be able to connect to storage1 at all. Disabling public access is part of the solution but is not sufficient on its own to route the connection through a private IP. Therefore, this option is insufficient by itself. 2. Option B: On VNet1, create a new subnet. - While creating a new subnet in VNet1 could be useful for isolating resources or for segmentation, it is not necessary for providing private connectivity between VM1 and storage1. The requirement is for VM1 to connect to storage1 using a private IP address, which can be achieved without creating a new subnet. This option does not directly address the need for private connectivity, so it is not the best choice. 3. Option C: For stora...

Author: VenomousSerpent42 · Last updated Aug 15, 2026