Python Text Processing with NLTK 2.0 Cookbook review

In the age where we have more information than we can possibly handle, we are seeing NLP being use in more and more software. To get started into learning and using NLP, adopting NLTK as your training ground would be an excellent choice, given the vast number of documentations and discussions around the net, and Python being a language that is very easy to understand and work with.

I’ve picked up Python Text Processing with NLTK Cookbook 2.0 to check out what are the problems that can be solve using NLP and NLTK. As a disclosure, I’ve work with Python in the past but I have little experience and knowledge with NLP. I’ve also gone through the O’Reilly NLTK book few months ago, but haven’t really used it to solve real world problems.

The Review

Python Text Processing with NLTK is written in a recipe style in which each little sections defines what problems we’re trying to solve, how to prepare the “ingredients” – things you need to install, and actually working on the problem, complete with code examples and explanations.

If a piece of software needs installation, the author would point us to websites that guides the installation process, which is nice as it cuts out unnecessary noise.

The author make use of up-to-date tools like MongoDB and Redis in his recipes to solve problem when necessary. A nice bonus since these tools are being adopt in more areas and situations nowadays, and knowing how it can be used to your advantage with NLP/NLTK is a huge plus.

It is possible to jump to a certain section and start working on the recipes in that particular section, but that requires you to have some NLP knowledge, as some terms used in a particular recipe might have been explained in the earlier recipes, in that case the author will refer you to the related chapters.

I like it where in each recipe, the author provides extra knowledge on the particular problem, like how a problem can be enhance and solve in another way, or what we need to do if the problem on hand changed, and some extra technical tips, which is very nice and useful.

Conclusion

Overall the book is very easy to follow. If you’re thinking about the O’Reilly’s NLTK book – Natural Language Processing with Python, IMHO this book and the O’Reilly NLTK book complements each other. The O’Reilly NLTK book focuses more on getting you to know NLP and the features and usage of NLTK , while Python Text Processing with NLTK teaches us how we would implement NLP/NLTK with tools like MongoDB into solving real world problems. It is suitable for busy programmers and NLP reserchers who doesn’t like to read lots of text to get their feet wet and start solving problems.

Resources

• The official site of Python Text Processing with NLTK 2.0 Cookbook
• The table of content of the Python Text Processing with NLTK 2.0 Cookbook.
• A free sample chapter – Chapter 3 – Creating Custom Corpora
• The author’s website

A friendly disclosure: get the book through the Amazon link banner above if my review helps you, thanks!