I have slugged my way through my share of major texts and academic papers on machine learning, data mining, NLP, and so on, and this book is a breath of fresh air. It is the first book I've seen that makes the topic accessible and has the reader immediately work with a great cross section of useful methods and applications.
The writing is clear, and rests on a good selection of problems and applications, instead of highly technical descriptions of algorithms. It appears squarely aimed at people who want to learn by working with real problems and data, and build up a good, workable toolkit, in the process.
Within my own profession, technical writing, structured documentation and technologies like DITA and document reuse represent the cutting edge in controlled documentation but the profession has no answer to democratic writing and publishing, represented by the web and wikis, that are overwhelming controlled writing and publishing. This book finally opens that door and provides the profession with the means to work with democratic material. I strongly urge members of my profession to study this book, and learn a different approach to how content can be detected, rated, and organized.
Tim Harvey
Text Wrestler, Google
|
