Natural Language Generation
Ehud Reiter
TL;DR
Ehud Reiter's Natural Language Generation provides a broad, concept-focused survey of NLG, spanning rule-based and neural approaches, evaluation, safety, and real-world applications. It foregrounds enduring principles over rapid model-innovation trends, and emphasizes requirements gathering, rigorous evaluation, and safety-driven deployment. The book links foundational data-to-text and text-to-text tasks to practical domains like weather, healthcare, journalism, and BI, offering guidance on when to deploy rule-based, ML-based, or hybrid NLG systems. Its insistence on principled evaluation, transparent design, and thoughtful integration with human workflows aims to shape robust, trustworthy applied NLG practice. Overall, the work serves as a bridge between historical NLG methodologies and contemporary, large-language-model-enabled practice, with concrete lessons for researchers and engineers building real-world systems.
Abstract
This book provides a broad overview of Natural Language Generation (NLG), including technology, user requirements, evaluation, and real-world applications. The focus is on concepts and insights which hopefully will remain relevant for many years, not on the latest LLM innovations. It draws on decades of work by the author and others on NLG. The book has the following chapters: Introduction to NLG; Rule-Based NLG; Machine Learning and Neural NLG; Requirements; Evaluation; Safety, Maintenance, and Testing; and Applications. All chapters include examples and anecdotes from the author's personal experiences, and end with a Further Reading section. The book should be especially useful to people working on applied NLG, including NLG researchers, people in other fields who want to use NLG, and commercial developers. It will not however be useful to people who want to understand the latest LLM technology. There is a companion site with more information at https://ehudreiter.com/book/