AI-900 Knowledge Mining 2 — Questions and Answers
Question 1: Which built-in cognitive skill enables Azure Cognitive Search to read text from images and scanned documents?
- Image Analysis skill
- Optical Character Recognition (OCR) skill (Correct answer)
- Text Merge skill
- Document Extraction skill
Correct answer: Optical Character Recognition (OCR) skill
The OCR skill uses computer vision to detect and extract printed and handwritten text from image files and scanned PDFs, making that content available for full-text search.
Question 2: Which built-in skill in Azure Cognitive Search identifies the language in which a document is written?
- Text Split skill
- Sentiment Analysis skill
- Language Detection skill (Correct answer)
- Named Entity Recognition skill
Correct answer: Language Detection skill
The Language Detection skill evaluates text input and returns a language code (such as 'en' for English), enabling downstream skills to process content in the correct language.
Question 3: In Azure Cognitive Search, what does setting a field attribute to 'Retrievable' allow?
- The field can be used to sort search results
- The field's content can be returned in search result responses (Correct answer)
- The field can be used to filter search results
- The field is included in the full-text search index
Correct answer: The field's content can be returned in search result responses
When a field is marked as Retrievable, its value is returned in the search API response, allowing the application to display the field's content to users.
Question 4: What is semantic search in Azure Cognitive Search?
- A search mode that only matches exact keyword strings
- A capability that uses AI to understand query intent and return contextually relevant results (Correct answer)
- A feature that translates queries between multiple languages automatically
- A method that searches only within structured, tabular data
Correct answer: A capability that uses AI to understand query intent and return contextually relevant results
Semantic search applies deep learning models to re-rank results based on the meaning and context of the query and documents, going beyond simple keyword matching.
Question 5: Which Azure AI service provides the underlying AI capabilities (such as language, vision, and speech) used in Azure Cognitive Search skillsets?
- Azure Machine Learning
- Azure Cognitive Services (Correct answer)
- Azure OpenAI Service
- Azure Bot Service
Correct answer: Azure Cognitive Services
Azure Cognitive Search built-in skills are backed by Azure Cognitive Services APIs (such as Text Analytics and Computer Vision), which provide the AI processing for enrichment.
Question 6: What is a projection in the context of an Azure Cognitive Search knowledge store?
- A security policy that restricts who can query the index
- A definition that specifies how enriched content is shaped and stored in the knowledge store (Correct answer)
- A mapping between source document fields and index fields
- A scheduled task that refreshes the index with updated data
Correct answer: A definition that specifies how enriched content is shaped and stored in the knowledge store
Projections define the structure and destination (table, object, or file) of enriched content written to the knowledge store, shaping data for specific downstream consumption patterns.
Question 7: Which field attribute in Azure Cognitive Search must be enabled to allow users to filter search results by a specific field value?
- Retrievable
- Searchable
- Facetable
- Filterable (Correct answer)
Correct answer: Filterable
Setting a field as Filterable allows it to be used in $filter expressions in search queries, enabling users to narrow results to documents matching specific field values.
