Courseiva

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

What is the difference between extractive summarization and abstractive summarization?

⚠ Common exam trap

Many candidates confuse the terms 'extractive' and 'abstractive' with other AI workloads (like image processing) or assume one is always superior, when in fact the key difference is whether the summary uses existing sentences or generates new text.

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

Extractive pulls existing sentences; abstractive generates new text capturing the meaning

Extractive summarization identifies and extracts the most important sentences directly from the source text, while abstractive summarization generates new sentences that capture the core meaning, often using natural language generation techniques. In Azure AI Language, extractive summarization returns a set of ranked sentences from the original document, whereas abstractive summarization produces a concise summary that may rephrase content. This distinction is fundamental to understanding how different NLP workloads handle text summarization tasks.

Answer analysis

Option-by-option breakdown

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

  • Extractive works on text; abstractive works on images

    Why it's wrong here

    Both extractive and abstractive summarization are text-based natural language processing tasks; neither operates on images. Extractive summarization works by selecting and concatenating important sentences or phrases directly from the source document, while abstractive summarization generates entirely new sentences that paraphrase the core ideas. The distinction is in how the summary is produced, not the input modality — Azure AI Language's document summarization, for example, expects text as input regardless of which of the two models is used.

  • Extractive pulls existing sentences; abstractive generates new text capturing the meaning

    Why this is correct

    Extractive summarization is a selection task: it identifies the most salient sentences or spans in the original document and copies them verbatim into the summary, preserving the source wording and structure. Abstractive summarization, in contrast, is a generation task: it reads the source text, builds a semantic understanding, and then produces new, condensed sentences that express the same meaning using different phrasing. This fundamental contrast between copying existing sentences and generating new text is exactly what differentiates the two approaches in Azure AI Language's summarization APIs.

  • Extractive is for long documents; abstractive is for short text

    Why it's wrong here

    The applicability of extractive versus abstractive summarization is not determined by document length; both methods can be applied to everything from a short news snippet to a multi-page report. The real differentiator lies in the output generation: extractive methods reuse existing sentences, whereas abstractive methods create novel summaries by synthesizing content. While some Azure AI implementation details, such as chunking for long documents, may affect handling of large inputs, these are practical engineering concerns rather than the conceptual definition of the two approaches.

  • Extractive summarization is always less accurate than abstractive

    Why it's wrong here

    There is no inherent rule that extractive summarization is always less accurate than abstractive; accuracy is highly context-dependent. Extractive summaries are usually more faithful to the source because they use exact wording, making them reliable for tasks that require preserving specific facts or legal language, but they can be less coherent or omit implied relationships. Abstractive summaries can be more fluent and concise, yet they introduce a risk of hallucination or subtle misinterpretation. In practice, Azure AI offers both modes because each has strengths depending on whether fidelity or readability matters more for a given scenario.

About these practice questions

Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. 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.