AI-900 Practice Test 43 — 15 Questions

Question 1

A manufacturer trains a model to detect defective parts on an assembly line. Only 2% of parts are defective. The model predicts 'non-defective' for all parts and achieves 98% accuracy. Which metric best reveals the model's inability to identify defective parts?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Recall. Recall (sensitivity) measures the proportion of actual defective parts correctly identified by the model. With 98% accuracy but zero true positives (since the model labels everything as non-defective), recall is 0%, which directly exposes the model's failure to detect any defective parts despite high accuracy.

Answer

Accuracy

Answer

Precision

Answer

F1 Score

Question 2

A hospital deploys an AI system to assist doctors in interpreting MRI scans. The system highlights the regions of interest and provides a numeric confidence score for its findings, along with a list of the image features that contributed to the diagnosis. Which responsible AI principle is being applied?

Accepted Answer

Transparency. The system provides a numeric confidence score and a list of image features that contributed to the diagnosis, which directly supports the principle of Transparency. Transparency in responsible AI requires that AI systems are understandable and that their decisions can be explained to users, enabling clinicians to interpret and trust the output.

Answer

Fairness

Answer

Privacy

Answer

Accountability

Question 3

What is 'object detection' in computer vision and how does it differ from image classification?

Accepted Answer

Object detection locates each object with a bounding box and class label; classification labels the whole image. Object detection goes beyond image classification by not only identifying the class of objects present but also localizing each one with a bounding box. In contrast, image classification assigns a single label to the entire image, regardless of how many objects are present. This distinction is fundamental in computer vision workloads on Azure, where Custom Vision and Computer Vision API offer both capabilities.

Answer

Object detection and image classification produce the same output — both label the entire image

Answer

Image classification processes images faster than object detection because it is simpler

Answer

Object detection only works on images with a single object; classification handles multiple objects

Question 4

A development team creates an AI chatbot for a hospital website that answers patient queries. The team scripts the AI to always respond with a disclaimer that it is not a substitute for professional medical advice. Additionally, they include a mechanism for users to report inaccurate responses, which are then reviewed by a human team. Which Microsoft responsible AI principle is most directly being implemented by the reporting and human review mechanism?

Accepted Answer

Accountability. The reporting and human review mechanism directly implements the Accountability principle, which requires that AI systems be designed with clear lines of responsibility and oversight. By allowing users to flag inaccuracies and having a human team review those reports, the organization takes ownership of the system's outputs and ensures corrective actions can be taken. This goes beyond mere transparency or reliability—it establishes a feedback loop where humans remain ultimately responsible for the AI's behavior.

Answer

Fairness

Answer

Reliability and safety

Answer

Transparency

Question 5

An online retailer wants to build a recommendation system that learns from user interactions. The system suggests a product, and if the user clicks it, it receives a positive reward; if ignored, a negative reward. Over time, the system learns to make better suggestions. Which type of machine learning best describes this approach?

Accepted Answer

Reinforcement learning. Reinforcement learning is correct because the system learns by interacting with its environment (user clicks) and receiving rewards (positive for clicks, negative for ignores) to maximize cumulative reward over time. This trial-and-error feedback loop, without explicit labeled data, is the hallmark of reinforcement learning.

Answer

Supervised learning

Answer

Unsupervised learning

Answer

Semi-supervised learning

Question 6

What is 'top_p' (nucleus sampling) in Azure OpenAI and how does it differ from temperature?

Accepted Answer

Restricting token selection to those whose cumulative probability reaches p — an alternative diversity control to temperature. Top_p (nucleus sampling) in Azure OpenAI controls diversity by selecting tokens from the smallest set whose cumulative probability exceeds the threshold p, rather than sampling from the full probability distribution. This differs from temperature, which scales the logits before the softmax to flatten or sharpen the distribution; top_p dynamically cuts off the long tail of low-probability tokens, providing an alternative method to control randomness without affecting the relative ranking of high-probability tokens.

Answer

The maximum percentage of the context window used for generating output

Answer

The probability threshold above which the model considers a response correct

Answer

A parameter setting the minimum confidence before the model outputs a response

Question 7

A restaurant chain wants to build a voice-powered ordering system for its drive-through. The system must understand when a user wants to place an order, modify an existing order, or cancel an order. It also needs to extract specific details like the menu item name and quantity from the user's speech. Which Azure AI Language feature should they use to handle both intent recognition and entity extraction?

Accepted Answer

Conversational Language Understanding (CLU). Conversational Language Understanding (CLU) is the correct choice because it is specifically designed to handle both intent recognition (e.g., 'place order', 'modify order', 'cancel order') and entity extraction (e.g., menu item name, quantity) from natural language utterances. CLU uses a pre-built or custom model to map user input to intents and extract detailed entities, making it ideal for a voice-powered ordering system that needs to understand complex commands.

Answer

Custom text classification

Answer

Key phrase extraction

Answer

Question answering

Question 8

A retail company wants to use Azure Computer Vision to monitor product availability on shelves. They need to detect the presence and location of any product (e.g., a box, a bottle) on a shelf image, but they do not need to identify the specific product brand or type. Which prebuilt Azure Computer Vision capability should they use?

Accepted Answer

Object detection. Object detection is the correct choice because it identifies and locates multiple objects within an image by drawing bounding boxes around each detected item. For monitoring product availability on shelves, the company needs to know both the presence and position of products (e.g., boxes, bottles) without identifying specific brands or types, which aligns exactly with object detection's capability to output class labels (e.g., 'product') and coordinates.

Answer

Image classification

Answer

Optical Character Recognition (OCR)

Answer

Semantic segmentation

Question 9

What is the difference between AI and conventional programming?

Accepted Answer

AI learns rules from data; conventional programming requires explicit rule specification. The fundamental distinction between AI and conventional programming lies in how rules are derived. In conventional programming, developers explicitly code every rule and logic path (e.g., if-then-else statements). In AI, particularly machine learning, the system learns patterns and rules directly from labeled or unlabeled data through training algorithms, without being explicitly programmed for each scenario. This enables AI to handle complex tasks like image recognition or natural language understanding where manual rule specification is impractical.

Answer

AI runs faster than conventional programs

Answer

AI can only work with images; conventional programming works with all data types

Answer

Conventional programming is more accurate than AI

Question 10

What is 'healthcare AI' and what capabilities does Azure provide for it?

Accepted Answer

AI for extracting medical entities, radiology insights, clinical trial matching, and patient analysis. Healthcare AI refers to AI solutions tailored for the healthcare industry, and Azure provides specific capabilities such as extracting medical entities (e.g., symptoms, medications) via Azure Health Bot and Text Analytics for Health, analyzing radiology images with Azure AI Vision, matching patients to clinical trials using Azure Cognitive Services, and performing patient analysis with Azure Machine Learning. These capabilities support clinical decision-making and operational efficiency without replacing doctors.

Answer

AI that gives patients direct medical advice as a substitute for doctors

Answer

A hospital management system for scheduling, billing, and patient record management

Answer

AI that monitors patients' vitals in real time using IoT medical devices

Question 11

A retail chain uses ceiling-mounted cameras to monitor shelf inventory. They need to identify and locate individual products (e.g., a specific brand of cereal) within an image and count how many are present. Which Azure Computer Vision capability should they use?

Accepted Answer

Object detection. Object detection is the correct capability because it not only identifies the presence of a specific product (e.g., a brand of cereal) within an image but also localizes each instance by drawing bounding boxes around them, enabling an accurate count. Image classification would only label the entire image as containing cereal without locating individual boxes, while OCR and semantic segmentation serve different purposes (text extraction and pixel-level labeling, respectively).

Answer

Image classification

Answer

Optical character recognition (OCR)

Answer

Semantic segmentation

Question 12

A company uses Azure OpenAI Service to generate creative product descriptions. They want to increase the randomness and variety of the generated outputs to produce more diverse suggestions. Which parameter should they increase?

Accepted Answer

Temperature. Temperature controls the randomness of the model's output by scaling the logits before applying the softmax function. Increasing temperature (e.g., from 0.7 to 1.0) flattens the probability distribution, making lower-probability tokens more likely to be chosen, which increases diversity and creativity in generated text.

Answer

Top_p

Answer

Frequency penalty

Answer

Presence penalty

Question 13

A company uses Azure OpenAI Service to generate marketing copy. They notice that sometimes the generated text contains repetitive phrases or gets stuck in loops. They want to reduce this behavior without changing the overall creativity of the model. Which parameter should they adjust?

Accepted Answer

Increase the frequency_penalty parameter.. Increasing the frequency_penalty parameter reduces the likelihood of the model repeating the same phrases by penalizing tokens that have already appeared in the generated text. This directly addresses the repetitive loops without altering the overall creativity, as frequency_penalty specifically targets token frequency rather than randomness or diversity.

Answer

Decrease the temperature parameter.

Answer

Increase the presence_penalty parameter.

Answer

Decrease the top_p parameter.

Question 14

A travel booking website wants to automatically identify famous landmarks (e.g., Eiffel Tower, Taj Mahal) in photos uploaded by users. They want to use a prebuilt Azure Computer Vision feature without custom training. Which capability should they use?

Accepted Answer

Domain-specific models (Landmark detection). Azure Computer Vision includes prebuilt domain-specific models for landmark detection that can identify famous landmarks like the Eiffel Tower or Taj Mahal without any custom training. This capability is specifically designed to recognize well-known structures from user-uploaded photos, making it the ideal choice for the travel booking website's requirement.

Answer

Image classification

Answer

Optical character recognition (OCR)

Answer

Object detection

Question 15

What is 'zero-shot prompting' and how does it work?

Accepted Answer

Asking the model to perform a task without any examples, relying on pre-trained knowledge. Zero-shot prompting refers to instructing a generative AI model to perform a task without providing any examples in the prompt. The model relies entirely on its pre-trained knowledge—gained from vast datasets during training—to interpret the instruction and generate a relevant response. This is a core capability of large language models (LLMs) like GPT-4, enabling them to generalize to unseen tasks without task-specific fine-tuning.

Answer

Running the model for zero seconds to test if the API connection works

Answer

Prompting the model to generate a response with zero errors or hallucinations

Answer

A technique that removes all instructions from the prompt to test raw model behaviour