AI-900 Practice Test 2 — 25 Questions

Question 1

What is 'model interpretability' and which Azure tool helps with it?

Accepted Answer

Understanding why a model makes specific predictions by identifying influential features — supported by Azure ML's Responsible AI dashboard. Model interpretability refers to the ability to understand and explain why a machine learning model makes specific predictions, typically by identifying which input features most influenced the output. Azure Machine Learning's Responsible AI dashboard directly supports this through built-in interpretability components like feature importance plots and error analysis, enabling developers to debug models and build trust. Option B correctly pairs the definition with the specific Azure tool that implements it.

Answer

Understanding what programming language a model was written in

Answer

Translating model documentation into multiple languages

Answer

Monitoring how quickly a model responds to prediction requests

Question 2

What is the Azure AI Vision background removal feature used for?

Accepted Answer

Automatically separating foreground subjects from the background in images. Azure AI Vision background removal is designed to automatically separate foreground subjects from the background in images, producing a mask or a cut-out of the primary object. This feature uses deep learning models to identify and isolate the main subject, enabling further processing like compositing or analysis without the background.

Answer

Blurring the background to create depth of field effects

Answer

Identifying what type of background (indoor/outdoor) is in an image

Answer

Replacing backgrounds in video calls

Question 3

What is 'sentiment analysis' at the opinion mining level vs. document level in Azure AI Language?

Accepted Answer

Document-level assigns one overall sentiment; opinion mining extracts per-aspect sentiments. In Azure AI Language, document-level sentiment analysis assigns a single overall sentiment (positive, negative, neutral, or mixed) to the entire document, while opinion mining (a feature of aspect-based sentiment analysis) extracts sentiments for specific aspects or targets within the text, such as 'service' or 'food' in a restaurant review. This allows for granular, per-aspect sentiment detection rather than a single aggregate score.

Answer

Document-level is more accurate; opinion mining is a faster but less precise approximation

Answer

Opinion mining works only on social media posts; document-level works on all text types

Answer

Document-level sentiment requires training data; opinion mining is pre-built

Question 4

What is 'Azure AI Vision's image vectorisation' and how does it enable image search?

Accepted Answer

Converting images to semantic embedding vectors for similarity-based search and retrieval. Azure AI Vision's image vectorisation converts images into semantic embedding vectors—numerical representations that capture the visual content and meaning of an image. These vectors enable similarity-based search by allowing the system to compare the vector of a query image against a database of pre-computed image vectors, returning the most visually or semantically similar results.

Answer

Converting image files to a vectorised (lossless) format like SVG for web use

Answer

Drawing vector graphics from a description of an image's contents

Answer

Optimising image file size by converting to the most efficient vector format

Question 5

A data scientist trains a regression model to predict energy consumption for a smart building. The model achieves very low error on the training data but performs significantly worse on a held-out validation set. Which technique would most directly address this problem?

Accepted Answer

Regularization. The model's low training error but high validation error indicates overfitting, where the model has memorized the training data rather than learning generalizable patterns. Regularization (e.g., L1 or L2) directly penalizes large coefficients, reducing model complexity and improving generalization to unseen data.

Answer

Feature engineering

Answer

Cross-validation

Answer

Hyperparameter tuning

Question 6

What does it mean for an ML model to 'generalize'?

Accepted Answer

The model's ability to perform well on new, unseen data by learning underlying patterns rather than memorizing training examples. Generalization in machine learning refers to the model's ability to accurately predict outcomes on new, unseen data by learning the true underlying patterns from the training data, rather than simply memorizing the training examples (overfitting). A model that generalizes well will maintain high performance on a validation or test dataset that was not used during training, which is a core requirement for deploying reliable ML solutions in Azure Machine Learning.

Answer

Making the model work for all programming languages and platforms

Answer

Making the model output descriptions in plain language for non-technical users

Answer

Training a model that works for all possible tasks without specialization

Question 7

What is 'cross-validation' and when should it be used in machine learning?

Accepted Answer

Dividing data into k folds and training k times to get a more reliable performance estimate. Cross-validation is a resampling technique used to evaluate machine learning models by partitioning the original dataset into k equal-sized folds. The model is trained on k-1 folds and validated on the remaining fold, repeating this process k times so each fold serves as the validation set once. This provides a more robust and less biased estimate of model performance compared to a single train-test split, especially when data is limited.

Answer

Validating that a model works correctly across different Azure regions

Answer

Comparing two different models' predictions on the same test set

Answer

Checking whether training labels are consistent across different human annotators

Question 8

What is reinforcement learning?

Accepted Answer

Training an agent through rewards and penalties in an interactive environment. Reinforcement learning is a machine learning paradigm where an agent learns to make decisions by interacting with an environment, receiving rewards for desirable actions and penalties for undesirable ones. This trial-and-error process allows the agent to develop an optimal policy over time, distinct from supervised or unsupervised learning. In Azure, this is exemplified by services like Azure Machine Learning's reinforcement learning capabilities or integration with platforms like Ray RLlib.

Answer

A type of supervised learning that uses labeled training data

Answer

A clustering technique that groups similar data automatically

Answer

Using previously trained models on new tasks

Question 9

What is 'celebrity recognition' in Azure AI Vision and what are its responsible AI limitations?

Accepted Answer

Recognising well-known public figures in images, with responsible AI access restrictions. Celebrity recognition in Azure AI Vision is a specialized feature that identifies well-known public figures (e.g., actors, politicians, athletes) in images. It is not a general-purpose facial identification service; instead, it relies on a curated dataset of public figures and is subject to responsible AI access restrictions, including limited availability and usage policies to prevent misuse.

Answer

Identifying any person by their face in a photograph using a global identity database

Answer

Automatically tagging images with the names of all people photographed at an event

Answer

A feature available to all Azure customers for identifying any person in any image

Question 10

What is 'conversation summarisation' in Azure AI Language?

Accepted Answer

Generating concise summaries of dialogues capturing key points, decisions, and action items. Conversation summarization in Azure AI Language is a prebuilt feature that uses extractive and abstractive summarization techniques to generate concise summaries of dialogues, capturing key points, decisions, and action items. It is designed specifically for multi-turn conversations (e.g., customer service chats, meeting transcripts) and outputs a structured summary, not just a count of messages.

Answer

Summarising how many messages were exchanged in a conversation

Answer

A tool for moderators to summarise flagged content for compliance review

Answer

Automatically creating FAQ articles from the most common chatbot conversations

Question 11

A data scientist is building a machine learning model to predict whether a credit card transaction is fraudulent or legitimate. The dataset contains 100,000 historical transactions, each labeled as 'fraudulent' or 'legitimate'. Which type of machine learning task should the data scientist use in Azure Machine Learning?

Accepted Answer

Binary classification. Binary classification is the correct choice because the prediction task involves distinguishing between exactly two mutually exclusive classes: 'fraudulent' and 'legitimate'. In Azure Machine Learning, binary classification algorithms (e.g., Two-Class Logistic Regression, Two-Class Boosted Decision Tree) are designed to output a probability score for one of two labels, making them ideal for this fraud detection scenario.

Answer

Regression

Answer

Multi-class classification

Answer

Clustering

Question 12

What does the Azure AI Vision 'Image Analysis' capability return when analyzing an image?

Accepted Answer

Descriptions, objects, tags, and other semantic information about the image content. Azure AI Vision's Image Analysis capability uses pre-trained deep learning models to extract rich semantic information from images, including human-readable descriptions, a list of detected objects with bounding boxes, and a set of relevant tags. This goes far beyond basic metadata, making option B correct because it accurately captures the breadth of semantic outputs the service provides.

Answer

Only the file size and dimensions of the image

Answer

Only a single category label for the entire image

Answer

A 3D point cloud of the scene

Question 13

A retail company wants to use security cameras to analyze customer flow. They need to detect when a person enters a specific store zone, count how many people are in that zone at any given time, and track the direction each person moves within the zone. Which Azure Computer Vision capability should they use?

Accepted Answer

Spatial Analysis. Spatial Analysis is the correct Azure Computer Vision capability because it is specifically designed to analyze video feeds from cameras to detect people, count them in defined zones, and track their movement direction. Unlike general object detection, Spatial Analysis provides the specialized functions for zone occupancy and person trajectory tracking required by the retail scenario.

Answer

Object detection

Answer

Optical Character Recognition (OCR)

Answer

Semantic segmentation

Question 14

What is the difference between entities and intents in conversational language understanding?

Accepted Answer

Intents represent the user's goal; entities are the specific pieces of information within the utterance. In conversational language understanding (CLU), intents represent the user's overall goal or desired action (e.g., 'BookFlight'), while entities are specific data points extracted from the utterance that provide context for that intent (e.g., 'New York' as a destination). This distinction is fundamental to natural language processing (NLP) on Azure, where intents map to actions and entities provide the parameters needed to fulfill those actions.

Answer

Intents are for text; entities are for speech recognition

Answer

Intents are predefined answers; entities are user questions

Answer

They are the same concept with different names for clarity

Question 15

What is 'feature importance' in Azure Machine Learning and how is it used?

Accepted Answer

Quantifying how much each input variable contributes to a model's predictions. Feature importance is a technique in Azure Machine Learning that quantifies the contribution of each input variable (feature) to a model's predictions. It is used to interpret model behavior, identify the most influential features, and validate that the model aligns with domain knowledge. This is critical for debugging, improving model performance, and ensuring regulatory compliance.

Answer

Ranking which ML project features (notebooks, experiments, pipelines) are most used by the team

Answer

Determining which model features (capabilities) are included in each Azure ML pricing tier

Answer

The priority order in which data preprocessing steps are applied before training

Question 16

A city government implements an AI system to analyze traffic camera feeds and predict congestion. The system is found to be less accurate for neighborhoods with lower-income populations because historical traffic data from those areas is sparse. Which Microsoft responsible AI principle is most directly relevant to address this issue?

Accepted Answer

Fairness. The system's reduced accuracy for lower-income neighborhoods due to sparse historical data is a direct fairness issue. Fairness in AI requires that systems perform equitably across different demographic groups, and this scenario describes a clear disparity in model performance based on socioeconomic factors. Addressing this would involve techniques like data augmentation, reweighting, or collecting more representative data to mitigate bias.

Answer

Transparency

Answer

Accountability

Answer

Privacy and security

Question 17

What is the 'dual-use' problem in AI and why is it relevant to responsible deployment?

Accepted Answer

The risk that AI capabilities designed for good can also be used for harmful purposes. The 'dual-use' problem in AI refers to the risk that a technology designed for beneficial purposes can also be misapplied for harmful ends. This is central to responsible deployment because it forces organizations to consider not only the intended use case but also potential misuse, such as facial recognition systems used for surveillance or generative AI creating disinformation. Addressing dual-use requires implementing safeguards like usage policies, access controls, and ethical review boards.

Answer

When an AI model is licensed for use by two different organisations simultaneously

Answer

Deploying the same AI model for both training and inference to reduce costs

Answer

Combining two AI models to achieve better results than either model alone

Question 18

What ethical consideration is MOST important when deploying AI systems for hiring decisions?

Accepted Answer

Auditing for and mitigating bias that could disadvantage protected demographic groups. The most critical ethical consideration in AI-driven hiring is fairness and non-discrimination. AI systems can inadvertently learn and amplify historical biases present in training data, leading to unfair outcomes for protected groups under laws like Title VII of the Civil Rights Act. Auditing for and mitigating bias ensures the AI model's decisions are equitable and legally compliant, which is a core principle of responsible AI.

Answer

Ensuring the AI processes applications as quickly as possible

Answer

Making the AI the final decision-maker for all candidates

Answer

Ensuring the AI is only deployed in large companies

Question 19

What is 'ROC-AUC' and when is it a better metric than accuracy for classification?

Accepted Answer

A threshold-agnostic metric that measures discrimination ability — better than accuracy for imbalanced classes. ROC-AUC (Receiver Operating Characteristic - Area Under the Curve) is a threshold-agnostic metric that measures a model's ability to discriminate between positive and negative classes across all possible classification thresholds. It is a better metric than accuracy when dealing with imbalanced classes because accuracy can be misleadingly high if the model simply predicts the majority class, whereas ROC-AUC evaluates the trade-off between true positive rate and false positive rate independently of class distribution.

Answer

ROC-AUC is always better than accuracy regardless of the use case

Answer

A metric specifically for measuring multi-class classification across more than two classes

Answer

An evaluation metric only applicable to models trained on Azure Machine Learning

Question 20

What is 'Microsoft Semantic Kernel' and how does it relate to Azure OpenAI?

Accepted Answer

Microsoft's open-source SDK for orchestrating LLMs with plugins, memory, and planning. Microsoft Semantic Kernel is an open-source SDK that enables developers to orchestrate large language models (LLMs) like Azure OpenAI by integrating plugins, memory, and planning capabilities. It abstracts the complexity of chaining AI calls, managing context, and executing multi-step tasks, making it a core tool for building generative AI workloads on Azure.

Answer

A low-level kernel module that optimises GPU utilisation for Azure OpenAI inference

Answer

A tool for evaluating the semantic accuracy of Azure OpenAI model responses

Answer

Microsoft's proprietary alternative to Azure OpenAI for internal use only