Customer Reviews: Read 1 more reviews...
Good overview of topic, but assumes you must learn Python November 30, 2008 This book provides an good collection of the various algorithms that can be used in this brave new world of Web 2.0. However, I have found it difficult to use as it has all its code written in Python, a language I know little about. It seems to me that if the algorithms were explained in detail beforehand, rather than having to be worked out from the code, this would be a superb book.
Given the type of analysis this book uses, the language could be either Java (as that pervades everywhere), PHP (the language of server scripting), or (even better) Lisp, which Python attempts to emulate.
It talks about making use of the various APIs that are now available, letting you access data that can be manipulated. But it does not show how the data looks before the programs modify them, so if you want (as I do) to use a different language, you cannot see from what you need to modify, only the result.
Don't get me wrong, this is a good book, but the Python code is poorly laid out (never heard of spaces? And if you have to use spaces to indent, try three spaces), and for those like me who don't read Python it was difficult to see what is going on. I don't want to learn Python just to learn the algorithms. I would, perhaps perversely, be prepared to plough through Lisp code, and it would be good to see the Python code converted to Lisp (and Java for that matter).
So yes, this is an interesting book as it describes areas of interest, given the new types of data that are widely available, but I only give it three stars because it seems to be written for Python users, not the general programming audience.
Keeps coming back, again and again November 5, 2008 This is an excellent book that I keep coming back to again and again. It explains a variety of complex machine learning algorithms with easy to follow, clear, concise code. The only bad word I've heard about this book is that occasionally it would be useful to see the algorithms as equations as well as code, I didn't find that a problem at all though.
Very Useful October 22, 2008 I was a beginner in both python and general programming when I got this book together with a book on how to use Python in general.
I actually found this book more useful than the general Python book as the use of examples is a much better way of learning (in my opinion). The examples introduced get more and more complex through the book which then forces you to learn about those programs in order to understand what they've done.
The fact that it exposes you to live web based data from the beginning is very useful indeed. My only wish would be that the formatting would be a bit clear. Python seems to thrive on using indentation rather than brackets and it can be difficult to see exactly which level of indentation is being used.
You need a ruler to use it!
very good introduction September 11, 2008 This book really does an awesome job at guiding you through the process of using collective intelligence in your own applications. For me, a bit more explanation on the math would be nice, but if i just pay attention it still seems to stick, so it's not crucial. just, well, get it! it's good
Pete
A truly outstanding book March 29, 2008 7 out of 7 found this review helpful
This book is definitely worth having on your bookshelf. It breaks down one of the most complex and demanding subjects into delightful, succinct and digestible pieces.
Having dredged my way through similar books in the past, I wasn't expecting too much, but Programming Collective Intelligence is remarkable. I found this book gripping; completing each example was very rewarding, and has now got me hooked on learning more.
I ended up working my way through the entire book, and really feel that it was time well spent. I now feel that I actually understand the concepts and algorithms surrounding machine learning/AI/data mining, and as a bonus have become familiar with a huge number of web 2.0 APIs.
The source code is almost impeccable. Most programming books falter on sloppy and/or incomplete code, but Programming Collective Intelligence is well explained and has the complete Python code written in the book, as well as being available for download. I was able to do every example without any major problems, despite having never used the Python language before.
Well written & diagrammed, with good examples and wonderful explanations: this a fantastic book, and to be highly recommended.
|