Describe features of Natural Language Processing workloads on Azure →mediumMultiple ChoiceObjective-mapped
AI-900 Practice Question: Describe features of Natural Language Processing workloads on Azure
A legal research company needs to automatically extract specific case citation patterns (e.g., '123 U.S. 456') from thousands of legal documents. The standard named entity recognition in Azure AI Language does not recognize these custom citation formats. Which Azure AI Language feature should they use to create a model that extracts these custom entities?
⚠ Common exam trap
Many candidates confuse Custom Named Entity Recognition with Key Phrase Extraction, thinking that key phrases can capture structured patterns like citations, but Key Phrase Extraction only returns generic, unlabeled phrases and cannot be trained to recognize specific entity formats.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Custom Named Entity Recognition
Custom Named Entity Recognition (NER) is the correct choice because it allows you to train a model to extract domain-specific entities, such as legal citation patterns like '123 U.S. 456', that are not recognized by the pre-built NER in Azure AI Language. Unlike standard NER, which only identifies common entity types (e.g., person, location, date), Custom NER lets you define custom entity labels and train the model with labeled examples to recognize these specific patterns in legal documents.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Sentiment Analysis
Why it's wrong here
Sentiment analysis evaluates text for positive, negative, or neutral tone, producing a sentiment score and labels. It does not extract or identify specific entities like legal citations; its output contains no entity spans or structured data. Therefore, it is completely unsuitable for extracting structured legal references from documents.
When this WOULD be correct
A question asking which Azure AI Language feature determines the emotional tone of customer feedback (e.g., positive, negative, neutral) in product reviews would have Sentiment Analysis as the correct answer.
- ✗
Key Phrase Extraction
Why it's wrong here
Key Phrase Extraction in Azure AI Language returns a list of important phrases based on statistical salience, but it cannot be trained to recognize domain-specific entities like legal citations (e.g., 'Smith v. Jones, 123 F.3d 456'). It operates without custom training data, so it would extract generic keywords rather than the structured legal references the company needs.
- ✗
Conversational Language Understanding
Why it's wrong here
CLU is optimized for building task-oriented conversational apps that process user utterances into intents and entities. It is not designed for document-level entity extraction, and applying it to static legal documents would require treating each document as an utterance, which is inefficient and fails to handle lengthy, multi-entity texts effectively.
When this WOULD be correct
A company wants to build a chatbot that can understand legal queries (e.g., 'Find cases citing 123 U.S. 456') and extract the citation as an entity to trigger a search. CLU would be correct because it combines intent recognition and entity extraction for conversational interactions.
- ✓
Custom Named Entity Recognition
Why this is correct
Custom Named Entity Recognition lets you define your own entity types (e.g., case citations, statutes) and train a model by providing annotated examples of those entities in context. After training, the model can automatically extract the exact legal citations from documents, aligning with the requirement for custom entity extraction in a specialized domain.
Option-by-option analysis
Why each answer is right or wrong
Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The AI-900 exam frequently reuses these exact scenarios with slightly different constraints.
✓Custom Named Entity RecognitionCorrect answer▾
Why this is correct
Custom Named Entity Recognition lets you define your own entity types (e.g., case citations, statutes) and train a model by providing annotated examples of those entities in context. After training, the model can automatically extract the exact legal citations from documents, aligning with the requirement for custom entity extraction in a specialized domain.
✗Sentiment AnalysisWrong answer — click to see why▾
Why this is wrong here
Sentiment Analysis detects positive/negative sentiment in text, not custom entities like legal citations. It cannot be trained to recognize specific patterns such as '123 U.S. 456'.
★ When this WOULD be the correct answer
A question asking which Azure AI Language feature determines the emotional tone of customer feedback (e.g., positive, negative, neutral) in product reviews would have Sentiment Analysis as the correct answer.
Why candidates choose this
Candidates may confuse 'analysis' with 'extraction' and think Sentiment Analysis can be customized to extract any text pattern, not understanding it is limited to sentiment scoring.
✗Conversational Language UnderstandingWrong answer — click to see why▾
Why this is wrong here
Conversational Language Understanding (CLU) is designed for intent classification and entity extraction in conversational contexts (e.g., chatbots), not for extracting custom citation patterns from legal documents. The question requires custom entity extraction from text, which is handled by Custom Named Entity Recognition.
★ When this WOULD be the correct answer
A company wants to build a chatbot that can understand legal queries (e.g., 'Find cases citing 123 U.S. 456') and extract the citation as an entity to trigger a search. CLU would be correct because it combines intent recognition and entity extraction for conversational interactions.
Why candidates choose this
Candidates may confuse CLU's entity extraction capability with custom NER, not realizing CLU is optimized for conversational flows rather than batch document processing.
Analysis generated from the official AI-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”
Go deeper
Related to this question
Learn chapter
Types of AI Workloads
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
Key term
Azure AI Language
Azure AI Language is a cloud-based service from Microsoft that uses natural language processing to understand, analyze, and generate human language for applications.
About these practice questions
This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.