AI-900 Practice Exam 52 — 20 Questions

Question 1

A data scientist trains a machine learning model to predict housing prices. On the training data, the model achieves an R-squared value of 0.99, but on a separate validation dataset it achieves an R-squared of only 0.65. What is the most likely issue with this model?

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

Accepted Answer

Overfitting. The model performs exceptionally well on the training data (R² = 0.99) but poorly on the validation data (R² = 0.65), which is a classic symptom of overfitting. Overfitting occurs when the model learns noise and specific patterns in the training set that do not generalize to unseen data, often due to excessive complexity (e.g., too many features or deep decision trees). In Azure Machine Learning, this can be detected by comparing training and validation metrics in automated ML runs or by using regularization techniques like L1/L2 penalties.

Answer

Underfitting

Answer

High bias

Answer

Insufficient training data

Question 2

Which Azure AI service can read text from a photo of a street sign taken by a mobile device?

Accepted Answer

Azure AI Vision (Read API / OCR). Azure AI Vision's Read API (part of the Computer Vision service) is designed to extract printed and handwritten text from images, including photos of street signs. It uses optical character recognition (OCR) to detect and digitize text, making it the correct choice for reading text from a mobile device photo.

Answer

Azure AI Speech

Answer

Azure AI Language

Answer

Azure AI Translator

Question 3

What is 'model lineage' in Azure Machine Learning?

Accepted Answer

A tracked history of the dataset, code, hyperparameters, and compute used to produce a model. Model lineage in Azure Machine Learning is a tracked history that captures the complete lifecycle of a model, including the dataset, code, hyperparameters, and compute environment used to produce it. This is essential for reproducibility, auditability, and governance, as it allows data scientists to trace exactly how a model was trained and which artifacts were involved. Azure ML automatically logs this lineage through its run history and model registry, ensuring every model version is linked to its training run.

Answer

The family tree of model architectures showing which models inspired the design

Answer

The geographic lineage of training data showing which regions it was collected from

Answer

The sequence of model versions deployed to production over time

Question 4

What is 'imbalanced classification' handling using 'SMOTE'?

Accepted Answer

Generating synthetic minority class examples by interpolating between existing examples. SMOTE (Synthetic Minority Over-sampling Technique) is a data augmentation method that creates synthetic examples for the minority class by interpolating between existing minority class instances. It selects a minority example, finds its k-nearest neighbors from the same class, and generates new samples along the line segments connecting the example to those neighbors. This balances the class distribution without duplicating existing data or discarding majority class examples.

Answer

A technique for collecting more real minority class examples from external data sources

Answer

Removing majority class examples until all classes have equal representation

Answer

Setting model confidence thresholds to classify more examples as the minority class

Question 5

A company uses Azure OpenAI to build a customer service chatbot. They want to prevent malicious users from injecting prompts that cause the chatbot to behave unexpectedly, such as revealing its system instructions. Which responsible AI consideration is most directly relevant?

Accepted Answer

Reliability and Safety. Prompt injection attacks target the system by embedding malicious instructions in user input, causing the model to override its original directives or reveal sensitive information. This directly undermines the reliability and safety of the AI system, as the chatbot's behavior becomes unpredictable and potentially harmful. Azure OpenAI's safety systems (e.g., content filtering, abuse detection) are designed to mitigate such risks, making Reliability and Safety the most relevant responsible AI consideration.

Answer

Fairness

Answer

Privacy and Security

Answer

Inclusiveness

Question 6

What is 'image recognition for accessibility' and how does Microsoft's Seeing AI app use it?

Accepted Answer

An app using Azure AI Vision to describe scenes, read text, and identify objects aloud for blind users. Microsoft's Seeing AI app leverages Azure AI Vision (specifically the Computer Vision API) to perform real-time image recognition for accessibility. It describes scenes, reads text via optical character recognition (OCR), and identifies objects aloud, enabling blind or low-vision users to understand their surroundings through audio feedback.

Answer

A feature that makes AI models accessible to users without programming expertise

Answer

Accessibility compliance checking software that validates AI applications meet WCAG standards

Answer

Screen reader software that makes Azure portal accessible to keyboard-only users

Question 7

What is the purpose of Azure Machine Learning's dataset versioning?

Accepted Answer

Tracking changes to training data over time to enable reproducibility and auditing. Azure Machine Learning's dataset versioning allows data scientists to track changes to training data over time by creating immutable snapshots of datasets. This ensures reproducibility of experiments and provides an audit trail, which is critical for compliance and debugging model performance regressions.

Answer

Encrypting datasets with different security keys for each version

Answer

Creating multiple copies of training data in different storage regions

Answer

Limiting which team members can access different versions of training data

Question 8

A bank deploys an AI system to approve personal loans. The system uses a complex deep learning model that produces a decision (approve or reject) without any explanation of why. Loan applicants who are rejected are not given any reason. According to Microsoft's responsible AI principles, which principle is most directly violated by this system?

Accepted Answer

Transparency. The system's inability to provide any explanation for its loan approval or rejection decisions directly violates the transparency principle. Microsoft's responsible AI principle of transparency requires that AI systems be understandable and that users be informed about how decisions are made, including the factors that influenced the outcome. A black-box deep learning model that gives no reasoning or feedback to rejected applicants fails this requirement.

Answer

Fairness

Answer

Reliability and safety

Answer

Privacy and security

Question 9

A marketing team uses Azure OpenAI Service to generate headline ideas for a campaign. They find the generated headlines are often too similar and lack creativity. Which parameter should they increase to introduce more randomness in the generated text?

Accepted Answer

Temperature. (Temperature) is correct because temperature controls the randomness of token selection in the model's probability distribution. Increasing temperature (e.g., from 0.7 to 1.0) flattens the probability curve, making lower-probability tokens more likely to be chosen, which introduces more diversity and creativity in the generated headlines.

Answer

Frequency penalty

Answer

Top_p (nucleus sampling)

Answer

Presence penalty

Question 10

Drag and drop the steps to process text with Azure Text Analytics (Language service) into the correct order.

Accepted Answer

Step 1: Provision an Azure Text Analytics resource, then Step 2: Retrieve the endpoint and key, then Step 3: Instantiate the Text Analytics client, then Step 4: Submit a document for analysis and process the response.. Using Text Analytics involves setting up a resource, making an API call, and interpreting results.

Answer

Step 2: Retrieve the endpoint and key, then Step 1: Provision the resource, then Step 3: Instantiate the client, then Step 4: Submit a document.

Answer

Step 1: Provision the resource, then Step 3: Instantiate the client, then Step 2: Retrieve the endpoint and key, then Step 4: Submit a document.

Answer

Step 1: Provision the resource, then Step 2: Retrieve the endpoint and key, then Step 4: Submit a document, then Step 3: Instantiate the client.

Question 11

A data scientist trains a binary classification model to detect fraudulent transactions. The dataset contains 99% legitimate transactions (negative class) and 1% fraudulent transactions (positive class). The model predicts 'legitimate' for every transaction in the test set and achieves 99% accuracy. Which metric would best reveal that the model is failing to identify any fraudulent transactions?

Accepted Answer

Recall. Recall (also known as sensitivity or true positive rate) measures the proportion of actual positive cases (fraudulent transactions) that the model correctly identifies. With 99% accuracy but zero true positives, the recall is 0%, which immediately reveals the model's complete failure to detect fraud. In Azure Machine Learning, the classification metrics pane would show recall = 0.0 for the positive class, highlighting this issue despite high accuracy.

Answer

Accuracy

Answer

Precision

Answer

F1-score

Question 12

A legal firm needs to automatically extract key information from contracts, including the names of parties involved, important dates, and monetary amounts. Which Azure AI Language feature should they use to identify and extract these specific pieces of information from the text?

Accepted Answer

Named Entity Recognition (NER). Named Entity Recognition (NER) is the correct Azure AI Language feature because it is specifically designed to identify and categorize entities such as people (parties involved), dates, and monetary amounts from unstructured text. This directly matches the legal firm's requirement to extract key information from contracts.

Answer

Sentiment analysis

Answer

Key phrase extraction

Answer

Language detection

Question 13

A data scientist is training a regression model to predict house prices using features like square footage, number of bedrooms, and location. After evaluating the model on a test set, the data scientist wants to select a metric that measures the average magnitude of prediction errors in the same units as the target variable (price). Which evaluation metric should the data scientist use?

Accepted Answer

Root Mean Squared Error (RMSE). Root Mean Squared Error (RMSE) is the correct metric because it measures the average magnitude of prediction errors in the same units as the target variable (price). RMSE is computed as the square root of the average squared differences between predicted and actual values, which brings the error metric back to the original unit (e.g., dollars), making it directly interpretable for regression tasks like house price prediction.

Answer

Accuracy

Answer

F1 Score

Answer

Precision

Question 14

A developer uses Azure OpenAI Service to generate data transformation scripts. The generated scripts sometimes contain logical errors. To make the model's output more deterministic and reduce variability, which parameter should the developer decrease?

Accepted Answer

Temperature. Temperature controls the randomness of the model's output. Lowering temperature (e.g., from 0.7 to 0.1) makes the model more deterministic and focused, reducing variability and the likelihood of logical errors in generated scripts. This is the correct parameter to adjust for more consistent, less creative responses.

Answer

Top_p

Answer

Frequency penalty

Answer

Presence penalty

Question 15

A social media platform uses Azure OpenAI Service to generate summaries of user comments. The development team discovers that sometimes the generated summaries include offensive or harmful language that was present in the original comments. The team wants to ensure that the generated output is always free of hate speech, profanity, and self-harm references. What should the team configure in the Azure OpenAI Service?

Accepted Answer

Configure a content filter. Azure OpenAI Service provides built-in content filtering that can be configured to block hate speech, profanity, and self-harm references in both input prompts and generated completions. This ensures that even if offensive language appears in the original user comments, the generated summaries will be free of such harmful content. The content filter operates at the service level, applying predefined severity thresholds to filter out undesirable language.

Answer

Set the temperature parameter to 0

Answer

Increase the max_tokens parameter

Answer

Use a grounding source

Question 16

A customer service team wants to analyze thousands of support tickets to automatically categorize them into predefined topics like 'billing', 'technical issue', and 'account management'. They have a small set of labeled tickets for each category. Which Azure AI Language feature should they use?

Accepted Answer

Custom Text Classification. The Custom Text Classification feature of Azure AI Language allows you to train a model using a small set of labeled tickets to automatically categorize text into predefined topics like 'billing', 'technical issue', and 'account management'. This feature is specifically designed for scenarios where you have labeled data and need to classify documents into custom categories, making it ideal for analyzing support tickets.

Answer

Pre-built Text Classification

Answer

Sentiment Analysis

Answer

Key Phrase Extraction

Question 17

What is a prompt in the context of generative AI?

Accepted Answer

The input text or instruction given to a generative AI model to guide its output. In generative AI, a prompt is the input text or instruction provided to a model (such as GPT-4 or DALL-E) to guide its output. It acts as the starting context or query that the model uses to generate a relevant response, image, or completion. This is a fundamental concept in Azure OpenAI Service and other generative AI workloads, where prompt engineering is used to refine outputs.

Answer

A configuration file for training AI models

Answer

A reward signal used in reinforcement learning

Answer

A type of neural network activation function

Question 18

A company uses Azure OpenAI Service to generate marketing copy for social media posts. They want to prevent the model from producing content that contains offensive language, harmful stereotypes, or violent themes that go against their brand guidelines. Which feature should the company configure within Azure OpenAI Service?

Accepted Answer

Configuring the content filtering (responsible AI filters). B is correct because Azure OpenAI Service includes built-in content filtering (responsible AI filters) that automatically detects and blocks offensive language, harmful stereotypes, and violent themes in both input prompts and generated outputs. This feature enforces brand guidelines without requiring custom model modifications or manual oversight.

Answer

Fine-tuning the model with a custom dataset

Answer

Increasing the token limit per response

Answer

Using prompt engineering techniques

Question 19

What is 'feature engineering' and why does it matter for machine learning models?

Accepted Answer

Creating and transforming input variables using domain knowledge to improve model performance. Feature engineering is the process of creating new input variables or transforming existing ones using domain knowledge to help machine learning models better capture patterns in the data. It directly impacts model performance by making the underlying relationships more explicit, reducing noise, and enabling algorithms to learn more effectively. In Azure Machine Learning, this is often done through automated feature engineering tools or custom Python scripts within pipelines.

Answer

Building physical infrastructure features (GPU clusters) for model training

Answer

The process of selecting which machine learning algorithm to use for a task

Answer

Adding new computing nodes to a training cluster to speed up training

Question 20

What is 'energy and sustainability' as an AI application area?

Accepted Answer

Using AI to optimise energy grids, building efficiency, agriculture, and climate modelling. 'energy and sustainability' as an AI application area refers to using AI to solve environmental challenges, such as optimizing energy grids, improving building efficiency, enhancing agricultural yields, and advancing climate modeling. This aligns with Microsoft's definition of AI for sustainability, where AI models analyze data to reduce waste, predict energy demand, and support renewable integration. It is not about the energy cost of AI itself, but about applying AI to broader sustainability goals.

Answer

Measuring and reducing the energy consumed by AI model training itself

Answer

Powering AI data centres with 100% renewable energy sources

Answer

Creating AI models that require less energy to run than traditional algorithms