Courseiva

AI-900 Practice Question: Describe features of Natural Language Processing workloads on Azure

A medical transcription service wants to automatically identify and extract medical terms such as diagnoses, medications, and procedures from doctor's notes. The notes are unstructured text. They want to use a pre-trained Azure AI Language feature that can understand medical terminology. Which feature should they use?

⚠ Common exam trap

Test-takers frequently confuse general-purpose Key Phrase Extraction (which only pulls out high-level topics) with domain-specific NER for healthcare, which is the only option that can accurately identify and classify medical terms like diagnoses and medications without custom training.

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

Named Entity Recognition (NER) for healthcare

B is correct because the medical transcription service needs to extract specific medical entities (diagnoses, medications, procedures) from unstructured doctor's notes. Azure AI Language's Named Entity Recognition (NER) for healthcare is a pre-trained model specifically designed to identify and categorize medical terminology, including conditions, treatments, and medications, directly from clinical text without requiring custom training.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • Custom Text Classification

    Why it's wrong here

    Custom Text Classification requires you to define your own categories and train a model on labeled note samples before it can classify text. It is not a pre-built healthcare solution, so it would not automatically identify medical entities without extensive domain-specific labeling. This makes it a poor fit for a service that needs immediate, out-of-the-box extraction of medical terms.

  • Named Entity Recognition (NER) for healthcare

    Why this is correct

    Named Entity Recognition (NER) for healthcare is a pre-trained Azure AI Language capability specifically tuned to identify medical entities such as diagnoses, medications, procedures, and symptoms directly from free-text notes. It uses healthcare-specific models and ontologies to classify spans of text into structured medical concepts, requiring no custom training. This allows the transcription service to automatically extract relevant clinical information from each note.

  • Key Phrase Extraction

    Why it's wrong here

    Key Phrase Extraction returns the most salient general phrases in a document based on statistical patterns, but it does not understand medical semantics. For example, it might surface 'patient had chest pain' as a key phrase without classifying 'chest pain' as a symptom entity. It cannot reliably produce structured medical entities such as drug names or diagnoses, so it is unsuitable for this extraction task.

  • Sentiment Analysis

    Why it's wrong here

    Sentiment Analysis determines whether a piece of text expresses positive, negative, or neutral emotion by scoring opinion-bearing language. It does not segment or label medical entities at all, so it can tell you how a patient describes a condition but not identify what the condition is. This is fundamentally different from entity extraction and would not meet the transcription service's requirement to recognize medical concepts.

About these practice questions

One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

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.