AI Associate [AI-102] 2 — Questions and Answers
Question 1: Which Azure AI service provides pre-built models for extracting key-value pairs and tables from forms and documents?
- Azure Document Intelligence (Correct answer)
- Azure Computer Vision
- Azure Language Service
- Azure Custom Vision
Correct answer: Azure Document Intelligence
Azure Document Intelligence (formerly Form Recognizer) extracts key-value pairs, tables, and structured data from documents using pre-built and custom models.
Question 2: When configuring an Azure Cognitive Search index, what is the purpose of a 'suggester'?
- It indexes documents for full-text search
- It enables autocomplete and search-as-you-type functionality (Correct answer)
- It scores documents by relevance
- It encrypts the index at rest
Correct answer: It enables autocomplete and search-as-you-type functionality
A suggester in Azure Cognitive Search defines which fields support autocomplete and search suggestions as users type.
Question 3: In Azure Machine Learning, what does a 'pipeline' primarily enable?
- Real-time inference on streaming data
- Orchestrating multi-step ML workflows that can be reused and automated (Correct answer)
- Monitoring model drift in production
- Deploying models to IoT Edge devices
Correct answer: Orchestrating multi-step ML workflows that can be reused and automated
Azure ML pipelines orchestrate multi-step machine learning workflows—data prep, training, evaluation—as reusable, schedulable directed acyclic graphs.
Question 4: Which metric is most appropriate for evaluating a binary classification model on an imbalanced dataset?
- Accuracy
- F1 Score (Correct answer)
- Mean Squared Error
- R-squared
Correct answer: F1 Score
F1 Score balances precision and recall, making it more informative than accuracy when class distribution is skewed.
Question 5: What is the primary function of the Azure AI Content Safety service?
- Translating content between languages
- Detecting harmful or inappropriate content in text and images (Correct answer)
- Encrypting sensitive data in transit
- Generating content using large language models
Correct answer: Detecting harmful or inappropriate content in text and images
Azure AI Content Safety detects harmful, offensive, or inappropriate content across text and images to help moderate user-generated content.
Question 6: In the context of Azure OpenAI Service, what does 'temperature' control during text generation?
- The latency of the API response
- The maximum number of tokens generated
- The randomness and creativity of the model's output (Correct answer)
- The number of candidate responses to generate
Correct answer: The randomness and creativity of the model's output
Temperature controls output randomness: lower values produce more deterministic responses, while higher values increase creativity and variability.
Question 7: Which Azure service would you use to perform real-time speech-to-text transcription in a call center application?
- Azure Video Indexer
- Azure Speech Service (Correct answer)
- Azure Bot Service
- Azure Language Understanding
Correct answer: Azure Speech Service
Azure Speech Service provides real-time speech-to-text transcription with support for custom acoustic and language models suited for domain-specific vocabulary.
Which Azure AI service provides pre-built models for extracting key-value pairs and tables from forms and documents?