- Hardcover: 542 pages
- Publisher: CRC Press; 2 edition (31 January 2012)
- Language: English
- ISBN-10: 1439860912
- ISBN-13: 978-1439860915
- Product Dimensions: 15.9 x 3.2 x 24.1 cm
- Average Customer Review: Be the first to review this item
- Amazon Bestsellers Rank: #6,14,320 in Books (See Top 100 in Books)
Statistical and Machine-Learning Data Mining: Techniques for Better Predictive Modeling and Analysis of Big Data, Second Edition Hardcover – 31 Jan 2012
There is a newer edition of this item:
Customers who bought this item also bought
Customers who viewed this item also viewed
About the Author
Bruce Ratner, DM STAT-l Consulting
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter mobile phone number.
|5 star (0%)|
|4 star (0%)|
|3 star (0%)|
|2 star (0%)|
|1 star (0%)|
Most helpful customer reviews on Amazon.com
It is very repetitive. It would ~20% shorter if verbatim passages were removed. Many ideas are restated throughout.
It is a relatively expensive book. I expected more technical and complete coverage for the money.
It does not adequately cover algorithms, mathematical concepts, or how to completely run the analysis described.
It barely covers SAS implementation.
It is not well organized. It comes across as a series of lectures or blog posts, in contrast to being a coherent book.
a. I first learnt of him and his GenIQ software from the e-mails he was posting to a LinkedIn group on a daily basis.
b. Watch "Statistical modeling and analysis for database marketing" become "Statistical and machine-learning data mining" in its second edition - the first time I see a book go into (n+1) edition with a completely different title - adding three "hot" search terms: "data mining", "machine learning" and "big data". And it works (on some people) - the book has zilch to say about big data, yet here it is in the "Big data" section of a ("prestigious", according to Mr. Ratner) list published by IBM.
c. He posts a five-star rating of own book - anonymously, as "The Significant Statistician". Two more reviews are from fans, who earlier gave five stars to the first edition, and both decided to upgrade; in total, three of the seven five-star reviews (excluding Mr. Ratner's own) come from his home state, NY, which is an interesting geographical spike. With one exception, the positive reviews have 2-3 positive votes each; coincidentally, my own review got two negative votes on the day of posting; literally putting two and two (multiplied by seven? or eight?) together, I can make a guess about where the votes are coming from.
Mr. Ratner is also a bona fide statistics PhD, but it seems that in the decades after leaving grad school, he has not invested time in keeping his knowledge of the field up to date - or in writing his books, which are just another channel of GenIQ promotion. I have reviewed the first edition - take a look at the comments under that 2009 review - and am disappointed to see the second one just as poorly written (a half-page passage shows up, unchanged, three times - on pp. 18, 48, 90 - labor of love, you say?), poorly typeset and visually ugly, and, well, shallow.
If this is your first book on statistical and non-statistical methods of data analysis, you may well be impressed, but at $80, the wisdom is a tad overpriced, and why not get a proper book by a recognized author? (I recommend "Introduction to statistical learning" by James, Witten, Hastie and Tibshirani; "Doing data science" by Schutt and O'Neil is another, very different option). America has a proud tradition of garage inventors, but this one needs to spend more time in the garage.
This was recommend to me by my former mentor.
I am so glad I took his advice and bought this book.
I can say I have read 60% of the book and I plan to read the rest.