Summarization
Summarization is the process of condensing source content into a shorter representation that preserves its important information or meaning.
Starting Data Analytics with Generative AI and Python
Text summarization is the generation of a shorter text that preserves the important information and key ideas from a source text.
- How it is used
- Condenses negative customer reviews and longer business documents so readers can identify important points quickly.
- How deep it goes
- Defines summarization, contrasts extractive and abstractive choices, and demonstrates prompts and frequency-based preprocessing without developing model internals.
- Examples
- Compares a Portuguese delivery complaint, frequency-based Python output, and ChatGPT responses across five customer reviews.
- What you can do after
- Apply it via ChatGPT or Python NLTK, then check whether the output retains the original review’s important information.
Read how Starting Data Analytics with Generative AI and Python explains summarization →
Build AI-Enhanced Web Apps
Text summarization is the generation of a shorter representation that preserves the important information in source text.
- How it is used
- Turns uploaded documents or pasted text into concise summaries in a web application.
- How deep it goes
- Works through loading, chunking, MapReduce invocation, UI integration, and parsing caveats rather than model-training internals.
- Examples
- Uses PDF and DOCX uploads, pasted text, a MapReduce chain invocation, and a summarization web-app UI.
- What you can do after
- Build it from scratch with LangChain.js and a web UI that accepts documents or text and returns summaries.
Read how Build AI-Enhanced Web Apps explains summarization →
Data Analysis with LLMs
Text summarization is the production of a shorter text that preserves the most important information from a source text.
- How it is used
- Uses review summaries to make customer feedback easier to analyze for recurring product strengths and weaknesses.
- How deep it goes
- Introduces summarization as an instruction-driven text task; does not explain a summarization algorithm or model internals.
- Examples
- Uses a laptop review praising its screen and speed, criticizing its weight, and approving its price.
- What you can do after
- Apply it via the ChatGPT web interface by supplying text and a task description that specifies the desired summary.
Effective Conversational AI
A summary is a shortened representation of content that preserves its key highlights and details.
- How it is used
- Condenses chatbot conversations for human-agent handoffs and restates customer problems in personalized greetings.
- How deep it goes
- Works through transcript prompts, prose summaries, structured extraction, and testing; does not develop model internals.
- Examples
- Uses a claim-status conversation ending in “representative,” plus prompts that produce prose summaries and JSON claim identifiers.
- What you can do after
- Apply it via LLM prompts to prepare handoff summaries and extract specified details, then check outputs against transcripts.
Read how Effective Conversational AI explains summarization →
AI Agents and Applications
Summarization is the generation of a shorter representation that preserves the important information in source content.
- How it is used
- Condenses short texts, large document collections, and web research into shorter outputs that retain key information.
- How deep it goes
- Works from a length-specified prompt through method selection and LangChain refinement code, including intermediate outputs and processing trade-offs.
- Examples
- Uses a 30-word home-PC summary, a stepwise historical-document summary, and web-page summaries for an Astorga research report.
- What you can do after
- Apply it via LangChain to choose a workflow, refine document summaries, and assemble sourced web research into a report.
Read how AI Agents and Applications explains summarization →
Knowledge Graphs and LLMs in Action
Summarization is the process of condensing information into a shorter representation that preserves its most relevant factual content.
- How it is used
- Presents knowledge-graph query results as concise, context-rich explanations alongside visual graph exploration.
- How deep it goes
- Explains the summarization position and prompt inputs, including questions, Cypher queries, records, selections, and node properties.
- Examples
- Uses crime-investigation results, node properties, vehicle details, and a partial license plate as concrete summary content.
- What you can do after
- Apply it via a result-aware prompt to produce concise summaries that emphasize important facts hidden within graph data.
Read how Knowledge Graphs and LLMs in Action explains summarization →
Microsoft Azure in Action
Summarization is the generation of a shorter representation that preserves the main information in a source text.
- How it is used
- Uses summarization for book discovery and for fitting longer source text into a model’s context window.
- How deep it goes
- Explains the purpose through a suitcase analogy but does not work through a summary prompt or assess summary quality.
- Examples
- Uses generated book summaries for customer discovery and an oversized book or conversation as a context-limit example.
- What you can do after
- Evaluate/understand when a shorter text could aid book discovery or reduce input size for a model.
Lead Developer Career Guide
The technique of restating the main points of a message in condensed form.
- How it is used
- Condenses documents and conversations to orient readers, confirm understanding, and prevent project misunderstandings.
- How deep it goes
- Explains the role through documentation structure and active-listening practice, without detailing a formal summarization procedure.
- Examples
- Uses developer-documentation summaries and a project-manager feature discussion to demonstrate reader orientation and misunderstanding prevention.
- What you can do after
- Apply it to summarize documentation and repeat project requests clearly enough to reveal misunderstandings early.
Read how Lead Developer Career Guide explains summarization →
Spring AI in Action
Summarization is the technique of condensing lengthy text into a shorter representation that preserves its essential meaning.
- How it is used
- Applies generative AI to turn lengthy documents into concise, usable summaries such as board-game quick-start instructions.
- How deep it goes
- Demonstrates prompt design, ChatClient integration, typed results, document extraction, dependency setup, and an HTTP endpoint.
- Examples
- Uses board-game rules, a Burger Battle PDF upload, quick-start instructions, and a JSON response containing the summary and game title.
- What you can do after
- Apply it via Spring AI ChatClient to build an endpoint that extracts uploaded documents and returns concise typed summaries.
Build an AI Agent (From Scratch)
Summarization is a context-reduction technique that replaces information with a shorter representation of its meaning, typically generated by a language model.
- How it is used
- Reduces growing agent context within a run or across accumulated session history by condensing older conversation.
- How deep it goes
- Details target selection, preservation boundaries, structured prompts, state tracking, and fallback ordering.
- Examples
- Uses execution histories organized around key findings, tools called, current status, and remaining work.
- What you can do after
- Operate it by preserving the original question and recent messages, summarizing intermediate work, and tracking the last summarized position.
Read how Build an AI Agent (From Scratch) explains summarization →
Building LLM Applications with DSPy
Summarization is the task of producing a shorter representation that preserves the important information in a source text.
- How it is used
- Uses summarization as a language-model task and a case for developing more complex evaluation metrics.
- How deep it goes
- Explains summary-quality concerns and a basic DSPy signature, while noting that evaluation requires more than simple correctness checks.
- Examples
- Uses the opening 20 stanzas of The Rime of the Ancient Mariner and a DSPy poem-to-summary prediction.
- What you can do after
- Apply summarization through DSPy's Predict interface by passing a poem and receiving a summarization field.
Read how Building LLM Applications with DSPy explains summarization →
Building Reliable AI Systems
The process of condensing information into a shorter representation that preserves important content.
- How it is used
- Compresses long-running agent conversations into running memory while retaining key context.
- How deep it goes
- Shows dynamic summary updates and a basic LLM-backed function, but not production safeguards.
- Examples
- Uses a chatbot conversation and a Python example summarizing hiking-related messages.
- What you can do after
- Apply via LangChain to summarize conversation messages and update an existing summary.
Read how Building Reliable AI Systems explains summarization →
Related concepts
- language modelused by · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
- extractive summarizationincludes · Extractive summarization produces a shorter representation of source content by retaining selected original words, phrases, sentences, or passages and omitting other material.
- abstractive summarizationincludes · Abstractive summarization produces a concise account of source information using newly composed wording rather than only selecting text from the source.
- promptuses · An input, typically containing instructions or context, supplied to an artificial-intelligence model to guide or elicit its output.
- ROUGEmeasured by · ROUGE is a family of automatic text-generation evaluation metrics that compare generated text with reference text using lexical overlap, often emphasizing recall.
- ChatGPTused by · ChatGPT is OpenAI’s conversational AI service, which lets users interact with generative AI models and related tools through natural-language prompts.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.