AI-900 Practice Test 19 — 15 Questions

Question 1

What is 'customer churn prediction' as an AI workload and what ML type does it use?

Accepted Answer

Using supervised classification to predict which customers are likely to cancel or become inactive. Customer churn prediction is a supervised machine learning workload where historical customer data (e.g., usage patterns, support interactions, billing history) is used to train a classification model. The model learns to assign a binary label (churn or not churn) to new customers, making it a supervised classification task. This directly matches option B, which correctly identifies the use of supervised classification to predict likely churners.

Answer

Analysing customer complaints to identify the root cause of service dissatisfaction

Answer

Detecting when a customer has already churned based on their last login date

Answer

Using NLP to understand why customers write negative reviews before leaving

Question 2

A data scientist is training a model to predict whether a patient has a rare disease (1% prevalence). The model predicts 'no disease' for all patients and achieves 99% accuracy, but fails to identify any actual cases. Which metric would best reveal this failure?

Accepted Answer

Recall. Recall (sensitivity) measures the proportion of actual positive cases correctly identified. With 1% disease prevalence and a model that predicts 'no disease' for all patients, recall is 0% because zero true positives are found. Accuracy (99%) is misleading here because the model fails to detect any rare disease cases, and recall directly exposes this failure.

Answer

Precision

Answer

F1 score

Answer

Mean absolute error

Question 3

Which Azure AI service detects and identifies human faces in images, including attributes like age estimate and emotion?

Accepted Answer

Azure AI Face. Azure AI Face is the correct service because it is specifically designed to detect and identify human faces in images, and it can extract attributes such as age estimates, emotions (e.g., happiness, sadness), and facial landmarks. Unlike general-purpose image analysis, Azure AI Face uses specialized face detection models and returns face rectangles along with optional attribute data.

Answer

Azure AI Vision

Answer

Azure AI Custom Vision

Answer

Azure AI Video Indexer

Question 4

What is the purpose of splitting data into training, validation, and test sets in machine learning?

Accepted Answer

To evaluate model performance honestly on data it hasn't seen during training. Splitting data into training, validation, and test sets is essential for honestly evaluating a model's performance on unseen data. The training set teaches the model patterns, the validation set tunes hyperparameters and prevents overfitting, and the test set provides a final, unbiased estimate of how the model will perform on new, real-world data. This separation ensures that the model's accuracy metrics reflect its generalization ability rather than memorization of the training data.

Answer

To increase the total amount of data available for training

Answer

To make training faster by using smaller datasets

Answer

To comply with data privacy regulations

Question 5

What is 'prompt flow' in Azure AI Foundry?

Accepted Answer

A visual development tool for building, testing, and deploying LLM application pipelines. Prompt flow in Azure AI Foundry is a visual development tool that enables developers to design, test, and deploy end-to-end pipelines for large language model (LLM) applications. It provides a graph-based interface to orchestrate LLM calls, data processing, and custom logic, making it easier to build complex generative AI workflows without writing extensive code.

Answer

A tool for managing the queue of prompt requests sent to Azure OpenAI during peak usage

Answer

An automated system that suggests improvements to prompts based on output quality metrics

Answer

A monitoring dashboard showing the flow of prompts through an AI application in production

Question 6

A customer service organization has thousands of support tickets labeled with predefined categories such as 'Billing', 'Technical', and 'Account Management'. They want to build a solution that automatically assigns a category to new, incoming tickets. The categories are fixed and known in advance. Which Azure AI Language service feature should they use?

Accepted Answer

Custom Text Classification. Custom Text Classification (B) is correct because the organization has a fixed set of predefined categories and needs to classify new support tickets into those categories. This feature allows you to train a custom model using labeled examples of 'Billing', 'Technical', and 'Account Management' tickets, enabling automatic assignment of incoming tickets to the correct category.

Answer

Prebuilt Text Analytics

Answer

Language Understanding (LUIS)

Answer

Translator

Question 7

What is 'latent semantic analysis' (LSA) and how does it relate to modern NLP embeddings?

Accepted Answer

An early matrix factorisation method for finding latent semantic relationships — predecessor to neural embeddings. Latent Semantic Analysis (LSA) is an early matrix factorization technique that uses singular value decomposition (SVD) to reduce the dimensionality of a term-document matrix, revealing latent semantic relationships between words and documents. This approach is a direct predecessor to modern neural embeddings (e.g., Word2Vec, GloVe), which also learn low-dimensional vector representations of words based on co-occurrence patterns, but with deeper non-linear transformations.

Answer

A legal analysis technique for discovering hidden clauses in contracts

Answer

A technique for analysing the structure of sentences to detect grammatical errors

Answer

Latent Semantic Analysis is the same as Large Language Model analysis

Question 8

What is data drift in the context of deployed machine learning models?

Accepted Answer

When production data distribution changes from the training data distribution over time. Data drift refers to the phenomenon where the statistical properties of the input data a deployed model receives in production change over time, diverging from the distribution of the data used during training. This degradation can cause the model's predictions to become less accurate or unreliable, even if the model itself remains unchanged. In Azure Machine Learning, data drift is monitored using dataset monitors that compare production data distributions against the training baseline.

Answer

When training data is accidentally deleted from storage

Answer

When a model's weights change during inference

Answer

When data is moved between different Azure storage accounts

Question 9

What is 'semantic search' in Azure AI Search (cognitive search)?

Accepted Answer

Search that understands the meaning and intent of queries to return conceptually relevant results. Semantic search in Azure AI Search uses advanced AI models to understand the meaning and intent behind a user's query, rather than relying solely on keyword matching. It re-ranks search results based on conceptual relevance to the query, enabling the system to return results that are semantically related even if they don't contain the exact keywords. This is powered by Azure's deep learning models, including transformer-based language models, to capture the context and semantics of the search terms.

Answer

A search that finds all documents containing the exact keywords typed by the user

Answer

Searching for programming code by its semantic meaning in a code repository

Answer

Restricting search results to documents tagged with specific metadata labels

Question 10

A real estate company wants to create an application that automatically generates floor plans from photographs of rooms. The application needs to identify and delineate every pixel in the image that corresponds to walls, doors, windows, and furniture. Which Azure Computer Vision capability should the company use?

Accepted Answer

Semantic Segmentation. Semantic segmentation is the correct choice because it classifies every pixel in an image into predefined categories (e.g., walls, doors, windows, furniture), producing a pixel-level mask. This is exactly what the application needs to delineate each structural element and object in the room photograph, enabling accurate floor plan generation.

Answer

Object Detection

Answer

Image Classification

Answer

Optical Character Recognition (OCR)

Question 11

What is 'wildlife monitoring' as a computer vision application and what Azure services power it?

Accepted Answer

Using computer vision to identify species, count populations, and track animals from camera trap images. 'wildlife monitoring' in the context of computer vision specifically refers to using AI to automatically analyze camera trap images to identify species, count populations, and track animal movements. Azure services such as Custom Vision (for training species-specific classifiers) and Computer Vision (for image analysis) power this by processing images captured in the field, enabling conservationists to gather data without manual review.

Answer

CCTV monitoring of wildlife parks to ensure visitor safety from animal encounters

Answer

Real-time video monitoring of endangered animal exhibits in zoos for welfare compliance

Answer

AI-powered smart thermostats that monitor and adapt wildlife sanctuary temperatures

Question 12

An autonomous vehicle system needs to both read the speed limit text on traffic signs and detect the presence and location of pedestrians crossing the road. Which combination of Azure Computer Vision capabilities should be used?

Accepted Answer

Optical Character Recognition (OCR) and Object Detection. The autonomous vehicle system requires two distinct capabilities: reading text from speed limit signs (OCR) and detecting the presence and location of pedestrians (Object Detection). OCR extracts text from images, while Object Detection identifies objects and provides bounding boxes around them, making option C the correct combination.

Answer

Image Classification and OCR

Answer

Semantic Segmentation and OCR

Answer

Face Detection and OCR

Question 13

What is 'model evaluation' and what metrics are used for different ML task types?

Accepted Answer

Different tasks use different metrics: F1 for classification, RMSE for regression, mAP for detection. Model evaluation is the process of assessing how well a trained machine learning model performs on unseen data. Different ML task types require different metrics because they measure distinct aspects of performance: for classification tasks, F1-score balances precision and recall; for regression tasks, RMSE (Root Mean Squared Error) quantifies prediction error in the same units as the target; for object detection tasks, mAP (mean Average Precision) evaluates both localization and classification accuracy. Option B correctly identifies these task-specific metrics.

Answer

Accuracy is the only metric needed for all ML task types

Answer

Model evaluation is only needed before deployment, not after

Answer

The only reliable evaluation is user feedback after the model is deployed in production

Question 14

What does Azure AI Speech's 'custom neural voice' capability allow organizations to do?

Accepted Answer

Create a unique branded synthetic voice trained on recordings of a specific speaker. Azure AI Speech's 'custom neural voice' capability allows organizations to create a unique, branded synthetic voice by training a neural text-to-speech model on recordings of a specific speaker. This enables high-quality, natural-sounding voice personalization for applications like virtual assistants, audiobooks, and customer service bots, while requiring explicit speaker consent and adherence to responsible AI guidelines.

Answer

Train speech recognition to understand unique industry vocabulary

Answer

Translate speech from one language to another in real time

Answer

Automatically identify accents and dialects in speech

Question 15

What does it mean for an AI system to be 'inclusive' according to Microsoft's responsible AI principles?

Accepted Answer

AI systems should empower all people including those with disabilities and from diverse backgrounds. Microsoft's responsible AI principle of inclusiveness requires that AI systems are designed to empower everyone, including people with disabilities and those from diverse cultural, linguistic, and socioeconomic backgrounds. This means the system should account for accessibility needs (e.g., screen readers, voice input) and avoid biases that could exclude or disadvantage any group.

Answer

AI systems should include as many features as possible regardless of user needs

Answer

AI data should include examples from every country in the world

Answer

All employees should be included in AI model training decisions