# Text Summarization

> Making a short version of a long text that keeps the main points.

**Field:** Artificial Intelligence  
**Category:** NLP & Language  
**Also known as:** Automatic Summarization, Summarisation

## Think of it like

Think of it like you tell a friend a whole film in three short lines.

## Definition

Text summarization makes a short version of a longer text that keeps its main points. There are two ways to do it. Extractive summaries copy the most important sentences as they are. Abstractive summaries write new, shorter sentences, the way a person would. Most modern language models write the second kind. These read better, but they can slip in things the text never said.

## Example

Jack gets a 40-page report before a meeting. A summary tool gives him one page with the five main findings.

## Why it matters

People have more to read than time to read it. A good summary tells them what is worth the full read.

## Common mistake

Trusting a written summary without a check. It can add a detail that sounds right but is not in the text.

## Three principles

- **Short and faithful.** Keeps the main points.
- **Extractive.** Copies the key sentences.
- **Abstractive.** Writes new, shorter ones.

## Related terms

- [Natural Language Processing (NLP)](https://learn.benibix.com/term/ai/natural-language-processing)
- [Language Model](https://learn.benibix.com/term/ai/language-model)
- [Machine Translation](https://learn.benibix.com/term/ai/machine-translation)
- [ROUGE Score](https://learn.benibix.com/term/ai/rouge-score)
- [Sequence-to-Sequence Model (Seq2Seq)](https://learn.benibix.com/term/ai/sequence-to-sequence-model)

![A long text of six lines on the left. One summary copies two of its lines as they are. The other writes one new, shorter line.](https://learn.benibix.com/diagrams/nlp-summarization.svg)

---

Source: https://learn.benibix.com/term/ai/text-summarization
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-18
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
