Big Data/Analytics Zone is brought to you in partnership with:

Andrew has posted 1 posts at DZone. View Full User Profile

Exploring Data With RapidMiner by Andrew Chisholm

02.07.2014
| 6609 views |
  • submit to reddit
Published by: PACKT
ISBN: 1782169334

Reviewer Ratings

Relevance:
5

Readability:
5

Overall:
5

Buy it now

One Minute Bottom Line

Andrew Chisholm shares tons of practical advices for data practitioners, using RapidMiner for data exploration. An excellent book for any data professional who wants to get the best out of RapidMiner and the data.

Review

If you’ve picked this book expecting that it explains what RapidMiner is and how to work with RapidMiner Studio GUI - put this book aside! It has none of it. If you hoped to learn how to build a stock market predictive model to beat the Wall Street– this book is not for you either.

The book “Exploring Data with RapidMiner” by Andrew Chisholm is for data practitioners who already have decent understanding and experience with RapidMiner. It covers the infamous “80% of project time” that data professionals spend on data understanding and preparation for data mining (the other 20% spent on the data mining itself).

So, before you start working on any serious project, especially if it includes millions of records, read this book. It includes tons of “tips and tricks” and “deep dives” to get the most out of RapidMiner and gives numerous practical advices on data exploration.

The following topics or notes were among my personal favorites:

  • advice that it’s often easier to modify XML than make changes in GUI for large datasets
  • a very detailed chapter of data visualization with great examples
  • usage of regular expressions, macros and debugging in RapidMiner
  • a discussion on windowing data (time series data transformation)

I strongly believe the book will find a respectful place on a bookshelf (whether physical or Kindle) for any serious data professional.

Published at DZone with permission of its author, Andrew Kuzmin.

(Note: Opinions expressed in this article and its replies are the opinions of their respective authors and not those of DZone, Inc.)