AI Associate [AI-102] 4 — Questions and Answers
Question 1: Which Azure AI Vision feature can you use to generate a natural-language description of an image automatically?
- Object detection
- Image captioning (Dense Captions) (Correct answer)
- Optical Character Recognition
- Custom image classification
Correct answer: Image captioning (Dense Captions)
Azure AI Vision's image captioning feature generates a human-readable natural-language description of an image's content.
Question 2: What is the key difference between 'system prompt' and 'user prompt' in an Azure OpenAI chat completion request?
- System prompts are billed separately; user prompts are free
- System prompts set assistant behavior and context; user prompts provide the actual user input (Correct answer)
- User prompts support function calling; system prompts do not
- System prompts are cached automatically; user prompts are not
Correct answer: System prompts set assistant behavior and context; user prompts provide the actual user input
The system prompt establishes the assistant's persona, constraints, and context, while the user prompt contains the actual conversational input from the user.
Question 3: A developer wants to detect whether a customer's message expresses frustration. Which Azure AI Language feature is most appropriate?
- Language detection
- Sentiment analysis with opinion mining (Correct answer)
- Text summarization
- Custom text classification
Correct answer: Sentiment analysis with opinion mining
Sentiment analysis with opinion mining evaluates the emotional tone of text and can identify negative sentiment indicative of frustration.
Question 4: When using Azure Bot Service with Direct Line channel, what format is used to exchange messages between the client and the bot?
- GraphQL queries
- Activity objects in JSON following the Bot Framework protocol (Correct answer)
- gRPC streams
- SOAP XML envelopes
Correct answer: Activity objects in JSON following the Bot Framework protocol
The Bot Framework protocol uses Activity objects serialized as JSON to represent messages, events, and other interactions over the Direct Line channel.
Question 5: Which feature of Azure AI Language allows you to build a model that classifies text into your own custom categories?
- Named Entity Recognition
- Custom Text Classification (Correct answer)
- Conversational Language Understanding
- Text Analytics for Health
Correct answer: Custom Text Classification
Custom Text Classification lets you train a model on labeled examples to categorize documents into categories you define.
Question 6: In responsible AI, which principle is violated when a facial recognition system performs significantly worse for darker-skinned individuals?
- Privacy and Security
- Fairness (Correct answer)
- Reliability and Safety
- Inclusiveness
Correct answer: Fairness
Disparate performance across demographic groups violates the Fairness principle, which requires AI systems to treat all people equitably.
Question 7: What is the purpose of 'semantic ranking' in Azure Cognitive Search?
- It replaces full-text keyword search with vector-only similarity
- It uses language model understanding to re-rank results by semantic relevance beyond keyword matching (Correct answer)
- It applies geospatial scoring to boost nearby results
- It enforces security trimming based on user roles
Correct answer: It uses language model understanding to re-rank results by semantic relevance beyond keyword matching
Semantic ranking applies a transformer-based model to re-rank the top keyword search results by their true semantic relevance to the query.
AI Associate [AI-102] 4
Which Azure AI Vision feature can you use to generate a natural-language description of an image automatically?