Easy, Practical, Great,
This review is from: Building Natural Language Generation Systems (Studies in Natural Language Processing) (Paperback)
As stated in the preface to this book, "This book describes NLG, which is a subfield of AI and CL that is concerned with building computer software systems that can produce meaningful texts in English or other human languages from some underlying information", and this is exactly what it does.
The heart of the book is contained in three chapters: 4) Document Planning, 5) Microplanning, and 6) Surface Realisation. It is in these three steps that the NLG pipeline is (usually) divided into. The other four chapters serve as an introduction to NLG, practical uses, and a description of the architectures followed by NLG systems.
Each chapter has a clear explanation, and a great list of references. It provides excellent information for the undergraduate or MSc student who is interested in Natural Language Processing and AI. I consider this book to be a great, easy read. It is not saturated with mathematical formulae which allows the reader to flow through the text easily.
Overall, I consider it a great read. It could do with a bit more depth in several sections, and examples using CCG (Possibly OpenCCG). However, considering it was written over 10 years ago, I still think it is pretty up to date regarding current practices.
(1 customer review)