Customer Reviews:
 Well researched, but a bit heavy on the maths. November 5, 2008
This book covers a lot of ground. The bibiography gets 5 stars, despite the fact that some of the URL's are already dead. I've posted a topic "URL related stuff" on the Manning Forum for the book. There is a good blend of case studies and code written in Java, harnessing things like Lucene, Nutch, WEKA, Java Data Mining. The book deals with ways of harvesting information from the Blogosphere as well as how to implement tag clouds and recommendation engines like Amazon's own system for books. It also deals with web crawling and how to do things like predictive modelling. It was this area which for me lost a star. I consider myself to be reasonably mathematically savvy, although some of this stuff I've not used in anger since my A Level days some 20+ years ago. But I found I couldn't keep up with some of chapters 9 & 10. I'll have to augment the book by referencing some of the citations. The way I'll endeavour to comprehend some of the code is running it through a debugger and re-reading the chapters again. It's not that the code is unclear. In fact I'd say the opposite. It's just the most practical way for me to get to grips with the heavy maths.
|
